Jean shorts raw denim Vice normcore, art party High Life PBR skateboard stumptown vinyl kitsch. Four loko meh 8-bit, tousled banh mi tilde forage Schlitz dreamcatcher twee 3 wolf moon. Chambray asymmetrical paleo salvia, sartorial umami four loko master cleanse drinking vinegar brunch. Pinterest DIY authentic Schlitz, hoodie Intelligentsia butcher trust fund brunch shabby chic Kickstarter forage flexitarian. Direct trade cold-pressed meggings stumptown plaid, pop-up taxidermy. Hoodie XOXO fingerstache scenester Echo Park. Plaid ugh Wes Anderson, freegan pug selvage fanny pack leggings pickled food truck DIY irony Banksy.
+
+
Hipster list
+
+
brunch
+
fixie
+
raybans
+
messenger bag
+
+
+
Hoodie Thundercats retro, tote bag 8-bit Godard craft beer gastropub. Truffaut Tumblr taxidermy, raw denim Kickstarter sartorial dreamcatcher. Quinoa chambray slow-carb salvia readymade, bicycle rights 90’s yr typewriter selfies letterpress cardigan vegan.
+
+
+
+
Pug heirloom High Life vinyl swag, single-origin coffee four dollar toast taxidermy reprehenderit fap distillery master cleanse locavore. Est anim sapiente leggings Brooklyn ea. Thundercats locavore excepteur veniam eiusmod. Raw denim Truffaut Schlitz, migas sapiente Portland VHS twee Bushwick Marfa typewriter retro id keytar.
+
+
+ We do not grow absolutely, chronologically. We grow sometimes in one dimension, and not in another, unevenly. We grow partially. We are relative. We are mature in one realm, childish in another.
+ —Anais Nin
+
+
+
Fap aliqua qui, scenester pug Echo Park polaroid irony shabby chic ex cardigan church-key Odd Future accusamus. Blog stumptown sartorial squid, gastropub duis aesthetic Truffaut vero. Pinterest tilde twee, odio mumblecore jean shorts lumbersexual.
This theme implements a built-in Jekyll feature, the use of Rouge, for syntax highlighting.
+It supports more than 100 languages.
+This example is in C++.
+All you have to do is wrap your code in a liquid tag:
This theme supports rendering beautiful math in inline and display modes using MathJax 3 engine. You just need to surround your math expression with $$, like $$ E = mc^2 $$. If you leave it inside a paragraph, it will produce an inline expression, just like \(E = mc^2\).
+
+
To use display mode, again surround your expression with $$ and place it as a separate paragraph. Here is an example:
an example of a distill-style blog post and main elements
+
+
+
+
+
+
NOTE:
+Citations, footnotes, and code blocks do not display correctly in the dark mode since distill does not support the dark mode by default.
+If you are interested in correctly adding dark mode support for distill, please open a discussion and let us know.
+
+
Equations
+
+
This theme supports rendering beautiful math in inline and display modes using MathJax 3 engine.
+You just need to surround your math expression with $$, like $$ E = mc^2 $$.
+If you leave it inside a paragraph, it will produce an inline expression, just like \(E = mc^2\).
+
+
To use display mode, again surround your expression with $$ and place it as a separate paragraph.
+Here is an example:
Citations are then used in the article body with the <d-cite> tag.
+The key attribute is a reference to the id provided in the bibliography.
+The key attribute can take multiple ids, separated by commas.
+
+
The citation is presented inline like this: (a number that displays more information on hover).
+If you have an appendix, a bibliography is automatically created and populated in it.
+
+
Distill chose a numerical inline citation style to improve readability of citation dense articles and because many of the benefits of longer citations are obviated by displaying more information on hover.
+However, we consider it good style to mention author last names if you discuss something at length and it fits into the flow well — the authors are human and it’s nice for them to have the community associate them with their work.
+
+
+
+
Footnotes
+
+
Just wrap the text you would like to show up in a footnote in a <d-footnote> tag.
+The number of the footnote will be automatically generated.This will become a hoverable footnote.
+
+
+
+
Code Blocks
+
+
Syntax highlighting is provided within <d-code> tags.
+An example of inline code snippets: <d-code language="html">let x = 10;</d-code>.
+For larger blocks of code, add a block attribute:
+
+
+ var x = 25;
+ function(x) {
+ return x * x;
+ }
+
+
+
Note:<d-code> blocks do not look well in the dark mode.
+You can always use the default code-highlight using the highlight liquid tag:
+
+
+
+
+
+
Layouts
+
+
The main text column is referred to as the body.
+It is the assumed layout of any direct descendants of the d-article element.
+
+
+
.l-body
+
+
+
For images you want to display a little larger, try .l-page:
+
+
+
.l-page
+
+
+
All of these have an outset variant if you want to poke out from the body text a little bit.
+For instance:
+
+
+
.l-body-outset
+
+
+
+
.l-page-outset
+
+
+
Occasionally you’ll want to use the full browser width.
+For this, use .l-screen.
+You can also inset the element a little from the edge of the browser by using the inset variant.
+
+
+
.l-screen
+
+
+
.l-screen-inset
+
+
+
The final layout is for marginalia, asides, and footnotes.
+It does not interrupt the normal flow of .l-body sized text except on mobile screen sizes.
+
+
+
.l-gutter
+
+
+
Emphasis, aka italics, with asterisks or underscores.
+
+
Strong emphasis, aka bold, with asterisks or underscores.
+
+
Combined emphasis with asterisks and underscores.
+
+
Strikethrough uses two tildes. Scratch this.
+
+
+
First ordered list item
+
Another item
+⋅⋅* Unordered sub-list.
+
Actual numbers don’t matter, just that it’s a number
+⋅⋅1. Ordered sub-list
+
And another item.
+
+
+
⋅⋅⋅You can have properly indented paragraphs within list items. Notice the blank line above, and the leading spaces (at least one, but we’ll use three here to also align the raw Markdown).
+
+
⋅⋅⋅To have a line break without a paragraph, you will need to use two trailing spaces.⋅⋅
+⋅⋅⋅Note that this line is separate, but within the same paragraph.⋅⋅
+⋅⋅⋅(This is contrary to the typical GFM line break behaviour, where trailing spaces are not required.)
URLs and URLs in angle brackets will automatically get turned into links.
+http://www.example.com or http://www.example.com and sometimes
+example.com (but not on Github, for example).
+
+
Some text to show that the reference links can follow later.
No language indicated, so no syntax highlighting.
+But let's throw in a <b>tag</b>.
+
+
+
Colons can be used to align columns.
+
+
+
+
+
Tables
+
Are
+
Cool
+
+
+
+
+
col 3 is
+
right-aligned
+
$1600
+
+
+
col 2 is
+
centered
+
$12
+
+
+
zebra stripes
+
are neat
+
$1
+
+
+
+
+
There must be at least 3 dashes separating each header cell.
+The outer pipes (|) are optional, and you don’t need to make the
+raw Markdown line up prettily. You can also use inline Markdown.
+
+
+
+
+
Markdown
+
Less
+
Pretty
+
+
+
+
+
Still
+
renders
+
nicely
+
+
+
1
+
2
+
3
+
+
+
+
+
+
Blockquotes are very handy in email to emulate reply text.
+This line is part of the same quote.
+
+
+
Quote break.
+
+
+
This is a very long line that will still be quoted properly when it wraps. Oh boy let’s keep writing to make sure this is long enough to actually wrap for everyone. Oh, you can putMarkdown into a blockquote.
+
+
+
Three or more…
+
+
+
+
Hyphens
+
+
+
+
Asterisks
+
+
+
+
Underscores
+
+
Here’s a line for us to start with.
+
+
This line is separated from the one above by two newlines, so it will be a separate paragraph.
+
+
This line is also a separate paragraph, but…
+This line is only separated by a single newline, so it’s a separate line in the same paragraph.
This theme supports generating various diagrams from a text description using jekyll-diagrams plugin.
+Below, we generate a few examples of such diagrams using languages such as mermaid, plantuml, vega-lite, etc.
+
+
Note: different diagram-generation packages require external dependencies to be installed on your machine.
+Also, be mindful of that because of diagram generation the fist time you build your Jekyll website after adding new diagrams will be SLOW.
+For any other details, please refer to jekyll-diagrams README.
+
+
Mermaid
+
+
Install mermaid using node.js package manager npm by running the following command:
+
npm install-g mermaid.cli
+
+
+
The diagram below was generated by the following code:
+
+
{% mermaid %}
+sequenceDiagram
+ participant John
+ participant Alice
+ Alice->>John: Hello John, how are you?
+ John-->>Alice: Great!
+{% endmermaid %}
+
Looks like there has been a mistake. Nothing exists here.
+
+
+
+
You will be redirected to the main page within 3 seconds. If not redirected, please click here.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/Gemfile b/Gemfile
deleted file mode 100644
index 9634030b..00000000
--- a/Gemfile
+++ /dev/null
@@ -1,16 +0,0 @@
-source 'https://rubygems.org'
-group :jekyll_plugins do
- gem 'jekyll'
- gem 'jekyll-diagrams'
- gem 'jekyll-email-protect'
- gem 'jekyll-feed'
- gem 'jekyll-github-metadata'
- gem 'jekyll-paginate-v2'
- gem 'jekyll-scholar'
- gem 'jekyll-sitemap'
- gem 'jekyll-target-blank'
- gem 'jekyll-twitter-plugin'
- gem 'jemoji'
- gem 'unicode_utils'
- gem 'webrick'
-end
diff --git a/_bibliography/dissertations.bib b/_bibliography/dissertations.bib
deleted file mode 100644
index cad63310..00000000
--- a/_bibliography/dissertations.bib
+++ /dev/null
@@ -1,329 +0,0 @@
----
----
-
-@phdthesis{wang2021toward,
- abbr={dissertation},
- title={Toward Robust Machine Learning by Countering Superficial Features},
- author={Wang, Haohan},
- year={2021},
- school={CMU},
- pdf={Thesis__Toward_Robust_Machine_Learning_by_Countering_Superficial_Features.pdf},
-}
-
-@phdthesis{al2021principles,
- abbr={dissertation},
- title={Principles of Learning in Multitask Settings: A Probabilistic Perspective},
- author={Al-Shedivat, Maruan},
- year={2021},
- school={CMU},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/alshedivat_phd_mld_2021_final.pdf},
-}
-
-@phdthesis{hu2021towards,
- abbr={dissertation},
- title={Towards Training AI Agents with All Types of Experiences: A Standardized ML Formalism},
- author={Hu, Zhiting},
- year={2021},
- school={CMU},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/phd_thesis_zhitinghu.pdf},
-}
-
-@phdthesis{Lee2021LearningEA,
- abbr={dissertation},
- title={Learning Embodied Agents with Scalably-Supervised Reinforcement Learning},
- author={Lisa Lee},
- year={2021},
- school={CMU},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/lslee_phd_mld_2021.pdf},
-}
-
-@phdthesis{qiao2021elastic,
- abbr={dissertation},
- title={Elastic Machine Learning Systems with Co-adaptation},
- author={Qiao, Aurick},
- year={2021},
- school={Carnegie Mellon University Pittsburgh, PA},
- pdf={https://kilthub.cmu.edu/articles/thesis/Elastic_Machine_Learning_Systems_with_Co-adaptation/16688761},
-}
-
-@phdthesis{lengerich2020sample,
- abbr={dissertation},
- title={Sample-Specific Models for Precision Medicine},
- author={Lengerich, Benjamin},
- year={2020},
- school={Carnegie Mellon University},
- pdf={https://kilthub.cmu.edu/articles/thesis/Sample-Specific_Models_for_Precision_Medicine/14460234},
-}
-
-@phdthesis{marchetti2020structured,
- abbr={dissertation},
- title={Structured Sparse Regression Methods for Learning from High-Dimensional Genomic Data},
- author={Marchetti-Bowick, Micol},
- year={2020},
- school={Carnegie Mellon University},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/mmarchet_phd_mld_2020.pdf},
-}
-
-@phdthesis{Zhang-2020-124839,
- abbr={dissertation},
- author = {Hao Zhang},
- title = {Machine Learning Parallelism Could Be Adaptive, Composable and Automated},
- year = {2020},
- school = {Carnegie Mellon University},
- pdf={http://www.cs.cmu.edu/%7Ehzhang2/files/hao_zhang_doctoral_dissertation.pdf},
-}
-
-@phdthesis{zheng2020learning,
- abbr={dissertation},
- title={Learning DAGs with Continuous Optimization},
- author={Zheng, Xun},
- year={2020},
- school={Carnegie Mellon University},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/thesis-zheng-xun.pdf},
-}
-
-@phdthesis{neiswangerpost,
- abbr={dissertation},
- title={Post-Inference Methods for Scalable Probabilistic Modeling and Sequential Decision Making},
- author={Neiswanger, Willie},
- year={2019},
- school={Carnegie Mellon University},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/cmu-ml-19-113-neiswanger.pdf},
-}
-
-@phdthesis{sachan2019towards,
- abbr={dissertation},
- title={Towards Literate Artificial Intelligence},
- author={Sachan, Mrinmaya},
- year={2019},
- school={Carnegie Mellon University},
- pdf={https://www.ml.cmu.edu/research/phd-dissertation-pdfs/cmu-ml-19-110-sachan.pdf},
-}
-
-@phdthesis{wei2019scheduling,
- abbr={dissertation},
- title={Scheduling for Efficient Large-Scale Machine Learning Training},
- author={Wei, Jinliang},
- year={2019},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/2019/CMU-CS-19-135.pdf},
-}
-
-
-
-
-@phdthesis{dai2018learning,
- abbr={dissertation},
- title={Learning with Staleness},
- author={Dai, Wei},
- year={2018},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2018/CMU-ML-18-100.pdf},
-}
-
-@phdthesis{kim2018framework,
- abbr={dissertation},
- title={Framework Design for Improving Computational Efficiency and Programming Productivity for Distributed Machine Learning},
- author={Kim, Jin Kyu},
- year={2018},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/2018/CMU-CS-18-127.pdf},
-}
-
-@phdthesis{xie2018diversity,
- abbr={dissertation},
- title={Diversity-promoting and Large-scale Machine Learning for Healthcare},
- author={Xie, Pengtao},
- year={2018},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2018/CMU-ML-18-106.pdf},
-}
-
-@phdthesis{parikh2015spectral,
- abbr={dissertation},
- title={Spectral Probabilistic Modeling and Applications to Natural Language Processing},
- author={Parikh, Ankur},
- year={2015},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2015/CMU-ML-15-102.pdf},
-}
-
-@phdthesis{leestructured,
- abbr={dissertation},
- title={Structured Sparse Models and Algorithms for Genetic Analaysis},
- author={Lee, Seunghak},
- year={2015},
- school={Carnegie Mellon University},
- website={http://reports-archive.adm.cs.cmu.edu/anon/2015/abstracts/15-100.html},
-}
-
-@phdthesis{zhao2013towards,
- abbr={dissertation},
- title={Towards Scalable Analysis of Images and Videos},
- author={Zhao, Bin},
- year={2014},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2014/CMU-ML-14-102.pdf},
-}
-
-@phdthesis{homodeling,
- abbr={dissertation},
- title={Modeling Large Social Networks in Context},
- author={Ho, Qirong},
- year={2014},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2014/CMU-ML-14-100.pdf},
-}
-
-@phdthesis{howrylak2013integrative,
- abbr={dissertation},
- title={An integrative computational framework for defining asthma endotypes},
- author={Howrylak, JA},
- year={2013},
- school={University of Pittsburgh},
- pdf={http://d-scholarship.pitt.edu/20305/1/thesis.pdf},
-}
-
-@phdthesis{kim2013reconstruction,
- abbr={dissertation},
- title={Reconstruction and Applications of Collective Storylines from Web Photo Collections},
- author={Kim, Gunhee},
- year={2013},
- school={Carnegie Mellon University},
- pdf={https://www.proquest.com/docview/1461727981?pq-origsite=gscholar&fromopenview=true},
-}
-
-
-@phdthesis{kolar2013uncovering,
- abbr={dissertation},
- title={Uncovering Structure in High-Dimensions: Networks and Multi-task Learning Problems},
- author={Kolar, Mladen},
- year={2013},
- school={Carnegie Mellon University Pittsburgh, PA},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2013/CMU-ML-13-106.pdf},
-}
-
-@phdthesis{puniyani2013spatiotemporal,
- abbr={dissertation},
- title={Spatiotemporal gene networks from ISH images},
- author={Puniyani, Kriti},
- year={2013},
- school={Carnegie Mellon University},
- pdf={https://www.lti.cs.cmu.edu/sites/default/files/research/thesis/2013/kriti_puniyani_spatiotemporal_gene_networks_from_ISH_images.pdf},
-}
-
-
-
-@phdthesis{martins2012geometry,
- abbr={dissertation},
- title={The geometry of constrained structured prediction: applications to inference and learning of natural language syntax},
- author={Martins, Andr{\'e} Filipe Torres},
- year={2012},
- school={Carnegie Mellon University},
- pdf={https://www.lti.cs.cmu.edu/sites/default/files/research/thesis/2012/andre_martins_the_geometry_of_constrained_structured_prediction_applications_to_inference_and_learning_of_natural_language_syntax.pdf},
-}
-
-@phdthesis{ray2012computational,
- abbr={dissertation},
- title={Computational Methods for Analyzing the Architecture and Evolution of the Regulatory Genome},
- author={Ray, Pradipta},
- year={2012},
- school={Carnegie Mellon University},
- pdf={https://personal.utdallas.edu/%7Epradiptaray/pubs/thesis_pradipta_12.pdf},
-}
-
-@phdthesis{shringarpure2012statistical,
- abbr={dissertation},
- title={Statistical Methods for studying Genetic Variation in Populations},
- author={Shringarpure, Suyash},
- year={2012},
- school={Carnegie Mellon University},
- pdf={https://kilthub.cmu.edu/articles/thesis/Statistical_Methods_for_studying_Genetic_Variation_in_Populations/6723164},
-}
-
-
-
-@phdthesis{ahmed2011modeling,
- abbr={dissertation},
- title={Modeling Content and Users: Structured Probabilistic Representation and Scalable Online Inference Algorithms},
- author={Ahmed, Amr},
- year={2011},
- school={Carnegie Mellon University},
- pdf={https://www.lti.cs.cmu.edu/sites/default/files/research/thesis/2011/amr_ahmed_modeling_content_and_users_structured_probabilistic_representation_and_scalable_online_inference_algorithms.pdf},
-}
-
-@phdthesis{curtis2011using,
- abbr={dissertation},
- title={Using Visualization and Automation to Accelerate Genetics Discovery},
- author={Curtis, Ross Eugene},
- year={2011},
- school={Carnegie Mellon University},
- pdf={https://cbd.cmu.edu/people/thesis/ross-e-curtis.pdf},
-}
-
-@misc{kamichetty2011structured,
- abbr={dissertation},
- title={Structured Probabilistic Models of Proteins across Spatial and Fitness Landscapes},
- author={Kamichetty, Hetunandan},
- year={2011},
- publisher={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/2011/CMU-CS-11-116.pdf},
-}
-
-@phdthesis{sohn2011learning,
- abbr={dissertation},
- title={Learning Ancestral Genetic Processes using Nonparametric Bayesian Models},
- author={Sohn, Kyung-Ah},
- year={2011},
- publisher={Carnegie Mellon University},
- pdf={https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.212.7705&rep=rep1&type=pdf},
-}
-
-@phdthesis{hanneke2009theoretical,
- abbr={dissertation},
- title={Theoretical foundations of active learning},
- author={Hanneke, Steve},
- year={2009},
- school={Carnegie Mellon University},
- pdf={http://reports-archive.adm.cs.cmu.edu/anon/ml2009/CMU-ML-09-106.pdf},
-}
-
-@phdthesis{zhao2007statistical,
- abbr={dissertation},
- title={Statistical alignment models for translational equivalence},
- author={Zhao, Bing},
- year={2007},
- school={Carnegie Mellon University},
- pdf={https://www.cs.cmu.edu/%7Ebzhao/Thesis.pdf},
-}
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
diff --git a/_bibliography/papers.bib b/_bibliography/papers.bib
deleted file mode 100644
index 7fac8cb3..00000000
--- a/_bibliography/papers.bib
+++ /dev/null
@@ -1,750 +0,0 @@
----
----
-
-@inproceedings{zhang2022ddg,
- abbr={CVPR},
- arxiv={https://arxiv.org/abs/2111.13839},
- code={https://github.com/hlzhang109/DDG},
- title={Towards Principled Disentanglement for Domain Generalization},
- author={Hanlin Zhang and Yi-Fan Zhang and Weiyang Liu and Adrian Weller and Bernhard Schölkopf and Eric P. Xing},
- booktitle={CVPR},
- year={2022}
-}
-
-@inproceedings{deng2021compression,
- abbr={NeurIPS},
- title={Multi-task Learning of Order-Consistent Causal Graphs},
- author={Xinshi Chen and Haoran Sun and Caleb Ellington and Eric Xing and Le Song},
- booktitle={Advances in Neural Information Processing Systems},
- year={2021}
-}
-
-@inproceedings{deng2021compression,
- abbr={EMNLP},
- arxiv={https://arxiv.org/abs/2109.06379},
- code={https://github.com/tanyuqian/ctc-gen-eval},
- title={Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation},
- author={Deng, Mingkai and Tan, Bowen and Liu, Zhengzhong and Xing, Eric P and Hu, Zhiting},
- booktitle={EMNLP},
- year={2021}
-}
-
-@inproceedings{yao2021knowledge,
- abbr={EMNLP},
- arxiv={https://arxiv.org/abs/2109.04707},
- title={Knowledge-Aware Meta-learning for Low-Resource Text Classification},
- author={Yao, Huaxiu and Wu, Yingxin and Al-Shedivat, Maruan and Xing, Eric P},
- booktitle={EMNLP},
- year={2021}
-}
-
-@article{tran2021computational,
- abbr={J. Chem. Phys.},
- title={Computational catalyst discovery: Active classification through myopic multiscale sampling},
- author={Tran, Kevin and Neiswanger, Willie and Broderick, Kirby and Xing, Eric and Schneider, Jeff and Ulissi, Zachary W},
- journal={The Journal of Chemical Physics},
- volume={154},
- number={12},
- pages={124118},
- year={2021},
- publisher={AIP Publishing LLC}
-}
-
-@article{wang2021coupled,
- abbr={BMC Bioinform},
- title={Coupled mixed model for joint genetic analysis of complex disorders with two independently collected data sets},
- author={Wang, Haohan and Pei, Fen and Vanyukov, Michael M and Bahar, Ivet and Wu, Wei and Xing, Eric P},
- journal={BMC bioinformatics},
- volume={22},
- number={1},
- pages={1--14},
- year={2021},
- publisher={BioMed Central}
-}
-
-@inproceedings{chen-etal-2021-geoqa,
- abbr={ACL-Findings},
- title = "{G}eo{QA}: A Geometric Question Answering Benchmark Towards Multimodal Numerical Reasoning",
- author = "Chen, Jiaqi and
- Tang, Jianheng and
- Qin, Jinghui and
- Liang, Xiaodan and
- Liu, Lingbo and
- Xing, Eric and
- Lin, Liang",
- booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
- month = aug,
- year = "2021",
- address = "Online",
- publisher = "Association for Computational Linguistics",
- url = "https://aclanthology.org/2021.findings-acl.46",
- doi = "10.18653/v1/2021.findings-acl.46",
- pages = "513--523",
-}
-
-@inproceedings{he2021towards,
- abbr={ACL},
- title={Towards Visual Question Answering on Pathology Images},
- author={He, Xuehai and Cai, Zhuo and Wei, Wenlan and Zhang, Yichen and Mou, Luntian and Xing, Eric and Xie, Pengtao},
- booktitle={Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)},
- pages={708--718},
- year={2021}
-}
-
-@inproceedings{zhou-etal-2021-generation,
- abbr={ACL},
- title = "On the Generation of Medical Dialogs for {COVID}-19",
- author = "Zhou, Meng and
- Li, Zechen and
- Tan, Bowen and
- Zeng, Guangtao and
- Yang, Wenmian and
- He, Xuehai and
- Ju, Zeqian and
- Chakravorty, Subrato and
- Chen, Shu and
- Yang, Xingyi and
- Zhang, Yichen and
- Wu, Qingyang and
- Yu, Zhou and
- Xu, Kun and
- Xing, Eric and
- Xie, Pengtao",
- booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)",
- month = aug,
- year = "2021",
- address = "Online",
- publisher = "Association for Computational Linguistics",
- url = "https://aclanthology.org/2021.acl-short.112",
- doi = "10.18653/v1/2021.acl-short.112",
- pages = "886--896",
- abstract = "Under the pandemic of COVID-19, people experiencing COVID19-related symptoms have a pressing need to consult doctors. Because of the shortage of medical professionals, many people cannot receive online consultations timely. To address this problem, we aim to develop a medical dialog system that can provide COVID19-related consultations. We collected two dialog datasets {--} CovidDialog {--} (in English and Chinese respectively) containing conversations between doctors and patients about COVID-19. While the largest of their kind, these two datasets are still relatively small compared with general-domain dialog datasets. Training complex dialog generation models on small datasets bears high risk of overfitting. To alleviate overfitting, we develop a multi-task learning approach, which regularizes the data-deficient dialog generation task with a masked token prediction task. Experiments on the CovidDialog datasets demonstrate the effectiveness of our approach. We perform both human evaluation and automatic evaluation of dialogs generated by our method. Results show that the generated responses are promising in being doctor-like, relevant to conversation history, clinically informative and correct. The code and the data are available at https://github.com/UCSD-AI4H/COVID-Dialogue.",
-}
-
-@inproceedings{tan2021progressive,
- abbr={ACL},
- title={Progressive Generation of Long Text with Pretrained Language Models},
- author={Tan, Bowen and Yang, Zichao and Al-Shedivat, Maruan and Xing, Eric and Hu, Zhiting},
- booktitle={Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
- pages={4313--4324},
- year={2021}
-}
-
-@inproceedings {273735,
-abbr={OSDI},
-author = {Aurick Qiao and Sang Keun Choe and Suhas Jayaram Subramanya and Willie Neiswanger and Qirong Ho and Hao Zhang and Gregory R. Ganger and Eric P. Xing},
-title = {Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning},
-booktitle = {15th {USENIX} Symposium on Operating Systems Design and Implementation ({OSDI} 21)},
-year = {2021},
-isbn = {978-1-939133-22-9},
-pages = {1--18},
-url = {https://www.usenix.org/conference/osdi21/presentation/qiao},
-publisher = {{USENIX} Association},
-month = jul,
-}
-
-@inproceedings{al2021federated,
- abbr={ICLR},
- title={Federated Learning via Posterior Averaging: A New Perspective and Practical Algorithms},
- author={Al-Shedivat, Maruan and Gillenwater, Jennifer and Xing, Eric and Rostamizadeh, Afshin},
- booktitle={International Conference on Learning Representations},
- year={2021}
-}
-
-@inproceedings{boecking2021interactive,
- abbr={ICLR},
- title={Interactive Weak Supervision: Learning Useful Heuristics for Data Labeling},
- author={Boecking, Benedikt and Neiswanger, Willie and Xing, Eric and Dubrawski, Artur},
- booktitle={International Conference on Learning Representations},
- year={2021}
-}
-
-@inproceedings{al2021data,
- abbr={AISTATS},
- title={On data efficiency of meta-learning},
- author={Al-Shedivat, Maruan and Li, Liam and Xing, Eric and Talwalkar, Ameet},
- booktitle={International Conference on Artificial Intelligence and Statistics},
- pages={1369--1377},
- year={2021},
- organization={PMLR}
-}
-
-@inproceedings{bang2021explaining,
- abbr={AAAI},
- title={Explaining A Black-box By Using A Deep Variational Information Bottleneck Approach},
- author={Bang, Seojin and Xie, Pengtao and Lee, Heewook and Wu, Wei and Xing, Eric},
- booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
- volume={35},
- number={13},
- pages={11396--11404},
- year={2021}
-}
-
-@article{zheng2020poly,
- abbr={PLoS Comput. Biol.},
- title={Poly (A)-DG: A deep-learning-based domain generalization method to identify cross-species Poly (A) signal without prior knowledge from target species},
- author={Zheng, Yumin and Wang, Haohan and Zhang, Yang and Gao, Xin and Xing, Eric P and Xu, Min},
- journal={PLoS computational biology},
- volume={16},
- number={11},
- pages={e1008297},
- year={2020},
- publisher={Public Library of Science San Francisco, CA USA}
-}
-
-@article{al2020contextual,
- abbr={JMLR},
- title={Contextual Explanation Networks.},
- author={Al-Shedivat, Maruan and Dubey, Avinava and Xing, Eric P},
- journal={J. Mach. Learn. Res.},
- volume={21},
- pages={194--1},
- year={2020}
-}
-
-@article{kandasamy2020tuning,
- abbr={JMLR},
- title={Tuning Hyperparameters without Grad Students: Scalable and Robust Bayesian Optimisation with Dragonfly.},
- author={Kandasamy, Kirthevasan and Vysyaraju, Karun Raju and Neiswanger, Willie and Paria, Biswajit and Collins, Christopher R and Schneider, Jeff and Poczos, Barnabas and Xing, Eric P},
- journal={J. Mach. Learn. Res.},
- volume={21},
- number={81},
- pages={1--27},
- year={2020}
-}
-
-@article{tran2020methods,
- abbr={Mach. Learn.: Sci. Technol.},
- title={Methods for comparing uncertainty quantifications for material property predictions},
- author={Tran, Kevin and Neiswanger, Willie and Yoon, Junwoong and Zhang, Qingyang and Xing, Eric and Ulissi, Zachary W},
- journal={Machine Learning: Science and Technology},
- volume={1},
- number={2},
- pages={025006},
- year={2020},
- publisher={IOP Publishing}
-}
-
-
-@article{kadambi2020wgan,
- abbr={IJCARS},
- title={WGAN domain adaptation for the joint optic disc-and-cup segmentation in fundus images},
- author={Kadambi, Shreya and Wang, Zeya and Xing, Eric},
- journal={International Journal of Computer Assisted Radiology and Surgery},
- volume={15},
- pages={1205--1213},
- year={2020},
- publisher={Springer}
-}
-
-@article{wang2020discovering,
- abbr={BMC Med. Genet.},
- title={Discovering weaker genetic associations guided by known associations},
- author={Wang, Haohan and Vanyukov, Michael M and Xing, Eric P and Wu, Wei},
- journal={BMC medical genomics},
- volume={13},
- number={3},
- pages={1--10},
- year={2020},
- publisher={Springer}
-}
-
-@article{zhang2020autosync,
- abbr={NeurIPS},
- title={AutoSync: Learning to Synchronize for Data-Parallel Distributed Deep Learning},
- author={Zhang, Hao and Li, Yuan and Deng, Zhijie and Liang, Xiaodan and Carin, Lawrence and Xing, Eric},
- journal={Advances in Neural Information Processing Systems},
- volume={33},
- year={2020}
-}
-
-@article{wu2020improving,
- abbr={NeurIPS},
- title={Improving GAN Training with Probability Ratio Clipping and Sample Reweighting},
- author={Wu, Yue and Zhou, Pan and Wilson, Andrew G and Xing, Eric and Hu, Zhiting},
- journal={Advances in Neural Information Processing Systems},
- volume={33},
- year={2020}
-}
-
-@article{plumb2020regularizing,
- abbr={NeurIPS},
- title={Regularizing black-box models for improved interpretability},
- author={Plumb, Gregory and Al-Shedivat, Maruan and Cabrera, {\'A}ngel Alexander and Perer, Adam and Xing, Eric and Talwalkar, Ameet},
- journal={Advances in Neural Information Processing Systems},
- volume={33},
- year={2020}
-}
-
-@inproceedings{tan2020summarizing,
- abbr={EMNLP},
- title={Summarizing Text on Any Aspects: A Knowledge-Informed Weakly-Supervised Approach},
- author={Tan, Bowen and Qin, Lianhui and Xing, Eric and Hu, Zhiting},
- booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
- pages={6301--6309},
- year={2020}
-}
-
-
-@inproceedings{lin-etal-2020-data,
- abbr={EMNLP Findings},
- title = "Data-to-Text Generation with Style Imitation",
- author = "Lin, Shuai and
- Wang, Wentao and
- Yang, Zichao and
- Liang, Xiaodan and
- Xu, Frank F. and
- Xing, Eric and
- Hu, Zhiting",
- booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2020",
- month = nov,
- year = "2020",
- address = "Online",
- publisher = "Association for Computational Linguistics",
- url = "https://aclanthology.org/2020.findings-emnlp.144",
- doi = "10.18653/v1/2020.findings-emnlp.144",
- pages = "1589--1598",
- abstract = "Recent neural approaches to data-to-text generation have mostly focused on improving content fidelity while lacking explicit control over writing styles (e.g., sentence structures, word choices). More traditional systems use templates to determine the realization of text. Yet manual or automatic construction of high-quality templates is difficult, and a template acting as hard constraints could harm content fidelity when it does not match the record perfectly. We study a new way of stylistic control by using existing sentences as {``}soft{''} templates. That is, a model learns to imitate the writing style of any given exemplar sentence, with automatic adaptions to faithfully describe the record. The problem is challenging due to the lack of parallel data. We develop a neural approach that includes a hybrid attention-copy mechanism, learns with weak supervisions, and is enhanced with a new content coverage constraint. We conduct experiments in restaurants and sports domains. Results show our approach achieves stronger performance than a range of comparison methods. Our approach balances well between content fidelity and style control given exemplars that match the records to varying degrees.",
-}
-
-@inproceedings{liu2020data,
- abbr={EMNLP Demo},
- title={A Data-Centric Framework for Composable NLP Workflows},
- author={Liu, Zhengzhong and Ding, Guanxiong and Bukkittu, Avinash and Gupta, Mansi and Gao, Pengzhi and Ahmed, Atif and Zhang, Shikun and Gao, Xin and Singhavi, Swapnil and Li, Linwei and others},
- booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations},
- pages={197--204},
- year={2020}
-}
-
-@inproceedings{huang2020self,
- abbr={ECCV},
- title={Self-challenging improves cross-domain generalization},
- author={Huang, Zeyi and Wang, Haohan and Xing, Eric P and Huang, Dong},
- booktitle={Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part II 16},
- pages={124--140},
- year={2020},
- organization={Springer}
-}
-
-@inproceedings{song2020generalized,
- abbr={IJCAI},
- title={Generalized Zero-Shot Text Classification for ICD Coding.},
- author={Song, Congzheng and Zhang, Shanghang and Sadoughi, Najmeh and Xie, Pengtao and Xing, Eric P},
- booktitle={IJCAI},
- pages={4018--4024},
- year={2020}
-}
-
-@inproceedings{dubey2020distributed,
- abbr={AISTATS},
- title={Distributed, partially collapsed MCMC for Bayesian nonparametrics},
- author={Dubey, Kumar Avinava and Zhang, Michael and Xing, Eric and Williamson, Sinead},
- booktitle={International Conference on Artificial Intelligence and Statistics},
- pages={3685--3695},
- year={2020},
- organization={PMLR}
-}
-
-@inproceedings{zheng2020learning,
- abbr={AISTATS},
- title={Learning sparse nonparametric dags},
- author={Zheng, Xun and Dan, Chen and Aragam, Bryon and Ravikumar, Pradeep and Xing, Eric},
- booktitle={International Conference on Artificial Intelligence and Statistics},
- pages={3414--3425},
- year={2020},
- organization={PMLR}
-}
-
-@inproceedings{korovina2020chembo,
- abbr={AISTATS},
- title={Chembo: Bayesian optimization of small organic molecules with synthesizable recommendations},
- author={Korovina, Ksenia and Xu, Sailun and Kandasamy, Kirthevasan and Neiswanger, Willie and Poczos, Barnabas and Schneider, Jeff and Xing, Eric},
- booktitle={International Conference on Artificial Intelligence and Statistics},
- pages={3393--3403},
- year={2020},
- organization={PMLR}
-}
-
-@inproceedings{wang2020high,
- abbr={CVPR},
- title={High-frequency component helps explain the generalization of convolutional neural networks},
- author={Wang, Haohan and Wu, Xindi and Huang, Zeyi and Xing, Eric P},
- booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
- pages={8684--8694},
- year={2020}
-}
-
-@inproceedings{ge2020supervised,
- abbr={RECOMB},
- title={Supervised Adversarial Alignment of Single-Cell RNA-seq Data},
- author={Ge, Songwei and Wang, Haohan and Alavi, Amir and Xing, Eric P and Bar-Joseph, Ziv},
- booktitle={RECOMB},
- year={2020}
-}
-
-@article{wang2019deep,
- abbr={BMC Bioinformatics},
- title={Deep mixed model for marginal epistasis detection and population stratification correction in genome-wide association studies},
- author={Wang, Haohan and Yue, Tianwei and Yang, Jingkang and Wu, Wei and Xing, Eric P},
- journal={BMC bioinformatics},
- volume={20},
- number={23},
- pages={1--11},
- year={2019},
- publisher={Springer}
-}
-
-@article{aragam2020identifiability,
- abbr={Annals of Statistics},
- title={Identifiability of nonparametric mixture models and bayes optimal clustering},
- author={Aragam, Bryon and Dan, Chen and Xing, Eric P and Ravikumar, Pradeep},
- journal={The Annals of Statistics},
- volume={48},
- number={4},
- pages={2277--2302},
- year={2020},
- publisher={Institute of Mathematical Statistics}
-}
-
-@article{marchetti2019penalized,
- abbr={Annals of Statistics},
- title={A penalized regression model for the joint estimation of eQTL associations and gene network structure},
- author={Marchetti-Bowick, Micol and Yu, Yaoliang and Wu, Wei and Xing, Eric P},
- journal={The Annals of Applied Statistics},
- volume={13},
- number={1},
- pages={248--270},
- year={2019},
- publisher={Institute of Mathematical Statistics}
-}
-
-@article{sachan-etal-2019-discourse,
- abbr={CL},
- title = "Discourse in Multimedia: A Case Study in Extracting Geometry Knowledge from Textbooks",
- author = "Sachan, Mrinmaya and
- Dubey, Avinava and
- Hovy, Eduard H. and
- Mitchell, Tom M. and
- Roth, Dan and
- Xing, Eric P.",
- journal = "Computational Linguistics",
- volume = "45",
- number = "4",
- month = dec,
- year = "2019",
- url = "https://aclanthology.org/J19-4002",
- doi = "10.1162/coli_a_00360",
- pages = "627--665",
- abstract = "To ensure readability, text is often written and presented with due formatting. These text formatting devices help the writer to effectively convey the narrative. At the same time, these help the readers pick up the structure of the discourse and comprehend the conveyed information. There have been a number of linguistic theories on discourse structure of text. However, these theories only consider unformatted text. Multimedia text contains rich formatting features that can be leveraged for various NLP tasks. In this article, we study some of these discourse features in multimedia text and what communicative function they fulfill in the context. As a case study, we use these features to harvest structured subject knowledge of geometry from textbooks. We conclude that the discourse and text layout features provide information that is complementary to lexical semantic information. Finally, we show that the harvested structured knowledge can be used to improve an existing solver for geometry problems, making it more accurate as well as more explainable.",
-}
-
-@article{kampffmeyer2018connnet,
- abbr={TIP},
- title={ConnNet: A long-range relation-aware pixel-connectivity network for salient segmentation},
- author={Kampffmeyer, Michael and Dong, Nanqing and Liang, Xiaodan and Zhang, Yujia and Xing, Eric P},
- journal={IEEE Transactions on Image Processing},
- volume={28},
- number={5},
- pages={2518--2529},
- year={2018},
- publisher={IEEE}
-}
-
-@article{lengerich2019learning,
- abbr={NeurIPS},
- title={Learning Sample-Specific Models with Low-Rank Personalized Regression},
- author={Lengerich, Ben and Aragam, Bryon and Xing, Eric P},
- journal={Advances in Neural Information Processing Systems},
- volume={32},
- pages={3575--3585},
- year={2019}
-}
-
-@article{wang2019learning,
- abbr={NeurIPS},
- title={Learning Robust Global Representations by Penalizing Local Predictive Power},
- author={Wang, Haohan and Ge, Songwei and Lipton, Zachary and Xing, Eric P},
- journal={Advances in Neural Information Processing Systems},
- volume={32},
- pages={10506--10518},
- year={2019}
-}
-
-@article{huang2019specific,
- abbr={NeurIPS},
- title={Specific and Shared Causal Relation Modeling and Mechanism-Based Clustering},
- author={Huang, Biwei and Zhang, Kun and Xie, Pengtao and Gong, Mingming and Xing, Eric P and Glymour, Clark},
- journal={Advances in Neural Information Processing Systems},
- volume={32},
- pages={13510--13521},
- year={2019}
-}
-
-@article{hu2019learning,
- abbr={NeurIPS},
- title={Learning Data Manipulation for Augmentation and Weighting},
- author={Hu, Zhiting and Tan, Bowen and Salakhutdinov, Russ R and Mitchell, Tom M and Xing, Eric P},
- journal={Advances in Neural Information Processing Systems},
- volume={32},
- pages={15764--15775},
- year={2019}
-}
-
-@inproceedings{hu2019texar,
- abbr={ACL Demo},
- title={Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation},
- author={Hu, Zhiting and Shi, Haoran and Tan, Bowen and Wang, Wentao and Yang, Zichao and Zhao, Tiancheng and He, Junxian and Qin, Lianhui and Wang, Di and Ma, Xuezhe and others},
- booktitle={Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations},
- pages={159--164},
- year={2019}
-}
-
-@inproceedings{jing2019show,
- abbr={ACL},
- title={Show, Describe and Conclude: On Exploiting the Structure Information of Chest X-ray Reports},
- author={Jing, Baoyu and Wang, Zeya and Xing, Eric},
- booktitle={Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics},
- pages={6570--6580},
- year={2019}
-}
-
-@inproceedings{qiao2019fault,
- abbr={ICML},
- title={Fault tolerance in iterative-convergent machine learning},
- author={Qiao, Aurick and Aragam, Bryon and Zhang, Bingjing and Xing, Eric},
- booktitle={International Conference on Machine Learning},
- pages={5220--5230},
- year={2019},
- organization={PMLR}
-}
-
-@inproceedings{zhang2019theoretically,
- abbr={ICML},
- title={Theoretically principled trade-off between robustness and accuracy},
- author={Zhang, Hongyang and Yu, Yaodong and Jiao, Jiantao and Xing, Eric and El Ghaoui, Laurent and Jordan, Michael},
- booktitle={International Conference on Machine Learning},
- pages={7472--7482},
- year={2019},
- organization={PMLR}
-}
-
-@inproceedings{xu2019multimodal,
- abbr={MLCH},
- title={Multimodal machine learning for automated ICD coding},
- author={Xu, Keyang and Lam, Mike and Pang, Jingzhi and Gao, Xin and Band, Charlotte and Mathur, Piyush and Papay, Frank and Khanna, Ashish K and Cywinski, Jacek B and Maheshwari, Kamal and others},
- booktitle={Machine Learning for Healthcare Conference},
- pages={197--215},
- year={2019},
- organization={PMLR}
-}
-
-@inproceedings{wang2019ellipse,
- abbr={ISBI},
- title={Ellipse detection of optic disc-and-cup boundary in fundus images},
- author={Wang, Zeya and Dong, Nanqing and Rosario, Sean D and Xu, Min and Xie, Pengtao and Xing, Eric P},
- booktitle={2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)},
- pages={601--604},
- year={2019},
- organization={IEEE}
-}
-
-@inproceedings{kim2019strads,
- abbr={ATC},
- title={Strads-ap: Simplifying distributed machine learning programming without introducing a new programming model},
- author={Kim, Jin Kyu and Aghayev, Abutalib and Gibson, Garth A and Xing, Eric P},
- booktitle={2019 $\{$USENIX$\}$ Annual Technical Conference ($\{$USENIX$\}$$\{$ATC$\}$ 19)},
- pages={207--222},
- year={2019}
-}
-
-@inproceedings{wei2019automating,
- abbr={EuroSys},
- title={Automating dependence-aware parallelization of machine learning training on distributed shared memory},
- author={Wei, Jinliang and Gibson, Garth A and Gibbons, Phillip B and Xing, Eric P},
- booktitle={Proceedings of the Fourteenth EuroSys Conference 2019},
- pages={1--17},
- year={2019}
-}
-
-@inproceedings{kampffmeyer2019rethinking,
- abbr={CVPR},
- title={Rethinking knowledge graph propagation for zero-shot learning},
- author={Kampffmeyer, Michael and Chen, Yinbo and Liang, Xiaodan and Wang, Hao and Zhang, Yujia and Xing, Eric P},
- booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
- pages={11487--11496},
- year={2019}
-}
-
-@inproceedings{dai2018toward,
- abbr={ICLR},
- title={Toward Understanding the Impact of Staleness in Distributed Machine Learning},
- author={Dai, Wei and Zhou, Yi and Dong, Nanqing and Zhang, Hao and Xing, Eric},
- booktitle={International Conference on Learning Representations},
- year={2018}
-}
-
-@inproceedings{wang2018learning,
- abbr={ICLR},
- title={Learning Robust Representations by Projecting Superficial Statistics Out},
- author={Wang, Haohan and He, Zexue and Lipton, Zachary C and Xing, Eric P},
- booktitle={International Conference on Learning Representations},
- year={2018}
-}
-
-@inproceedings{xu2018autoloss,
- abbr={ICLR},
- title={AutoLoss: Learning Discrete Schedule for Alternate Optimization},
- author={Xu, Haowen and Zhang, Hao and Hu, Zhiting and Liang, Xiaodan and Salakhutdinov, Ruslan and Xing, Eric},
- booktitle={International Conference on Learning Representations},
- year={2018}
-}
-
-@inproceedings{wang2019if,
- abbr={AAAI},
- title={What if we simply swap the two text fragments? a straightforward yet effective way to test the robustness of methods to confounding signals in nature language inference tasks},
- author={Wang, Haohan and Sun, Da and Xing, Eric P},
- booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
- volume={33},
- number={01},
- pages={7136--7143},
- year={2019}
-}
-
-@inproceedings{li2019knowledge,
- abbr={AAAI},
- title={Knowledge-driven encode, retrieve, paraphrase for medical image report generation},
- author={Li, Christy Y and Liang, Xiaodan and Hu, Zhiting and Xing, Eric P},
- booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
- volume={33},
- number={01},
- pages={6666--6673},
- year={2019}
-}
-
-@inproceedings{wang2019removing,
- abbr={PSB},
- title={Removing Confounding Factors Associated Weights in Deep Neural Networks Improves the Prediction Accuracy for Healthcare Applications},
- author={Wang, Haohan and Wu, Zhenglin and Xing, Eric P},
- booktitle={Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing},
- volume={24},
- pages={54},
- year={2019},
- organization={NIH Public Access}
-}
-
-@inproceedings{wang2019automatic,
- abbr={PSB},
- title={Automatic Human-like Mining and Constructing Reliable Genetic Association Database with Deep Reinforcement Learning},
- author={Wang, Haohan and Liu, Xiang and Tao, Yifeng and Ye, Wenting and Jin, Qiao and Cohen, William W and Xing, Eric P},
- booktitle={Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing},
- volume={24},
- pages={112},
- year={2019},
- organization={NIH Public Access}
-}
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-@string{aps = {American Physical Society,}}
-
-@book{einstein1956investigations,
- bibtex_show={true},
- title={Investigations on the Theory of the Brownian Movement},
- author={Einstein, Albert},
- year={1956},
- publisher={Courier Corporation,}
-}
-
-@article{einstein1950meaning,
- abbr={AJP},
- bibtex_show={true},
- title={The meaning of relativity},
- author={Einstein, Albert and Taub, AH},
- journal={American Journal of Physics,},
- volume={18},
- number={6},
- pages={403--404},
- year={1950},
- publisher={American Association of Physics Teachers,}
-}
-
-@article{PhysRev.47.777,
- abbr={PhysRev},
- title={Can Quantum-Mechanical Description of Physical Reality Be Considered Complete?},
- author={Einstein, A. and Podolsky, B. and Rosen, N.},
- abstract={In a complete theory there is an element corresponding to each element of reality. A sufficient condition for the reality of a physical quantity is the possibility of predicting it with certainty, without disturbing the system. In quantum mechanics in the case of two physical quantities described by non-commuting operators, the knowledge of one precludes the knowledge of the other. Then either (1) the description of reality given by the wave function in quantum mechanics is not complete or (2) these two quantities cannot have simultaneous reality. Consideration of the problem of making predictions concerning a system on the basis of measurements made on another system that had previously interacted with it leads to the result that if (1) is false then (2) is also false. One is thus led to conclude that the description of reality as given by a wave function is not complete.},
- journal={Phys. Rev.,},
- volume={47},
- issue={10},
- pages={777--780},
- numpages={0},
- year={1935},
- month={May},
- publisher=aps,
- doi={10.1103/PhysRev.47.777},
- url={http://link.aps.org/doi/10.1103/PhysRev.47.777},
- html={https://journals.aps.org/pr/abstract/10.1103/PhysRev.47.777},
- pdf={example_pdf.pdf},
- selected={true}
-}
-
-@article{einstein1905molekularkinetischen,
- title={{\"U}ber die von der molekularkinetischen Theorie der W{\"a}rme geforderte Bewegung von in ruhenden Fl{\"u}ssigkeiten suspendierten Teilchen},
- author={Einstein, A.},
- journal={Annalen der physik,},
- volume={322},
- number={8},
- pages={549--560},
- year={1905},
- publisher={Wiley Online Library}
-}
-
-@article{einstein1905movement,
- abbr={Ann. Phys.},
- title={Un the movement of small particles suspended in statiunary liquids required by the molecular-kinetic theory 0f heat},
- author={Einstein, A.},
- journal={Ann. Phys.,},
- volume={17},
- pages={549--560},
- year={1905}
-}
-
-@article{einstein1905electrodynamics,
- title={On the electrodynamics of moving bodies},
- author={Einstein, A.},
- year={1905}
-}
diff --git a/_bibliography/preprints.bib b/_bibliography/preprints.bib
deleted file mode 100644
index 14c676e7..00000000
--- a/_bibliography/preprints.bib
+++ /dev/null
@@ -1,134 +0,0 @@
----
----
-
-
-
-@article{zhang2022can,
- abbr={arXiv},
- title={Can Transformers be Strong Treatment Effect Estimators?},
- author={Zhang, Yi-Fan and Zhang, Hanlin and Lipton, Zachary C and Li, Li Erran and Xing, Eric P},
- journal={arXiv preprint arXiv:2202.01336},
- year={2022}
-}
-
-@article{wang2021toward,
- abbr={arXiv},
- title={Toward Learning Human-aligned Cross-domain Robust Models by Countering Misaligned Features},
- author={Wang, Haohan and Huang, Zeyi and Zhang, Hanlin and Xing, Eric},
- journal={arXiv preprint arXiv:2111.03740},
- year={2021}
-}
-
-@article{lengerich2021notmad,
- abbr={arXiv},
- title={NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters},
- author={Lengerich, Ben and Ellington, Caleb and Aragam, Bryon and Xing, Eric P and Kellis, Manolis},
- journal={arXiv preprint arXiv:2111.01104},
- year={2021}
-}
-
-@article{hu2021panoramic,
- abbr={arXiv},
- title={Panoramic Learning with A Standardized Machine Learning Formalism},
- author={Hu, Zhiting and Xing, Eric P},
- journal={arXiv preprint arXiv:2108.07783},
- year={2021}
-}
-
-@article{xiao2021amortized,
- abbr={arXiv},
- title={Amortized Auto-Tuning: Cost-Efficient Transfer Optimization for Hyperparameter Recommendation},
- author={Xiao, Yuxin and Xing, Eric P and Neiswanger, Willie},
- journal={arXiv preprint arXiv:2106.09179},
- year={2021}
-}
-
-@article{lengerich2020discriminative,
- abbr={medRxiv},
- title={Discriminative Subtyping of Lung Cancers from Histopathology Images via Contextual Deep Learning},
- author={Lengerich, Benjamin J and Al-Shedivat, Maruan and Alavi, Amir and Williams, Jennifer and Labbaki, Sami and Xing, Eric P},
- journal={medRxiv},
- year={2020},
- publisher={Cold Spring Harbor Laboratory Press}
-}
-
-@article{wang2020word,
- abbr={arXiv},
- title={Word shape matters: Robust machine translation with visual embedding},
- author={Wang, Haohan and Zhang, Peiyan and Xing, Eric P},
- journal={arXiv preprint arXiv:2010.09997},
- year={2020}
-}
-
-@article{zhang2020iterative,
- abbr={arXiv},
- title={Iterative graph self-distillation},
- author={Zhang, Hanlin and Lin, Shuai and Liu, Weiyang and Zhou, Pan and Tang, Jian and Liang, Xiaodan and Xing, Eric P},
- journal={arXiv preprint arXiv:2010.12609},
- year={2020}
-}
-
-@article{zou2020validate,
- abbr={arXiv},
- title={Validate and Enable Machine Learning in Industrial AI},
- author={Zou, Hongbo and Chen, Guangjing and Xie, Pengtao and Chen, Sean and He, Yongtian and Huang, Hochih and Nie, Zheng and Zhang, Hongbao and Bala, Tristan and Tulip, Kazi and others},
- journal={arXiv preprint arXiv:2012.09610},
- year={2020}
-}
-
-@article{wang2020squared,
- abbr={arXiv},
- title={Squared $$\backslash$ell\_2 $ Norm as Consistency Loss for Leveraging Augmented Data to Learn Robust and Invariant Representations},
- author={Wang, Haohan and Huang, Zeyi and Wu, Xindi and Xing, Eric P},
- journal={arXiv preprint arXiv:2011.13052},
- year={2020}
-}
-
-@article{dong2020towards,
- abbr={arXiv},
- title={Towards Robust Medical Image Segmentation on Small-Scale Data with Incomplete Labels},
- author={Dong, Nanqing and Kampffmeyer, Michael and Liang, Xiaodan and Xu, Min and Voiculescu, Irina and Xing, Eric P},
- journal={arXiv preprint arXiv:2011.14164},
- year={2020}
-}
-
-@article{lengerich2020disentangling,
- abbr={medRxiv},
- title={Disentangling Increased Testing From Covid-19 Epidemic Spread},
- author={Lengerich, Benjamin J and Neiswanger, Willie and Lengerich, Eugene J and Xing, Eric P},
- journal={medRxiv},
- year={2020},
- publisher={Cold Spring Harbor Laboratory Press}
-}
-
-
-@article{he2020sample,
- abbr={medRxiv},
- title={Sample-efficient deep learning for COVID-19 diagnosis based on CT scans},
- author={He, Xuehai and Yang, Xingyi and Zhang, Shanghang and Zhao, Jinyu and Zhang, Yichen and Xing, Eric and Xie, Pengtao},
- journal={medrxiv},
- year={2020},
- publisher={Cold Spring Harbor Laboratory Press}
-}
-
-@article{lengerich2020dropout,
- abbr={arXiv},
- title={On dropout, overfitting, and interaction effects in deep neural networks},
- author={Lengerich, Benjamin and Xing, Eric P and Caruana, Rich},
- journal={arXiv preprint arXiv:2007.00823},
- year={2020}
-}
-
-@article{neiswanger2019probo,
- abbr={arXiv},
- title={ProBO: Versatile Bayesian Optimization Using Any Probabilistic Programming Language},
- author={Neiswanger, Willie and Kandasamy, Kirthevasan and Poczos, Barnabas and Schneider, Jeff and Xing, Eric},
- journal={arXiv preprint arXiv:1901.11515},
- year={2019}
-}
-
-
-
-
-
-
diff --git a/_bibliography/talks.bib b/_bibliography/talks.bib
deleted file mode 100644
index 2719f0db..00000000
--- a/_bibliography/talks.bib
+++ /dev/null
@@ -1,52 +0,0 @@
----
----
-@article{hwang-ciai2022,
- abbr={CIAI},
- title = "On the Utility of Gradient Compression in Distributed Training Systems",
- year = 2022,
- journal = "CIAI Colloquium, MBZUAI",
- talk = "https://www.youtube.com/watch?v=gprhrinr3I4",
- abstract = "A rich body of prior work has highlighted the existence of communication bottlenecks in distributed training. To alleviate these bottlenecks, a long line of recent research proposes to use gradient compression methods. In this talk, Dr. Hongyi Wang (CMU) will first evaluate gradient compression methods' efficacy and compare their scalability with optimized implementations of synchronous data-parallel SGD across more than 200 realistic distributed setups. The observation is that, surprisingly, only in six cases out of 200, do gradient compression methods provide promising speedup. He will then introduce our extensive investigation to identify the root causes of this phenomenon and present a performance model that can be used to identify the benefits of gradient compression for a variety of system setups. Finally, Dr. Hongyi will propose a list of desirable properties (along with two algorithmic instances) that a gradient compression method should satisfy, in order for it to provide significant speedup in real distributed training systems."
-}
-
-@article{baidu2021,
- abbr={Baidu},
- title = "From Learning, to Meta-Learning, to "Lego-Learning” -- theory, system, and applications",
- year = 2021,
- journal = "Baidu",
- talk = "https://youtu.be/_12g9kpL4K8?t=1",
- abstract = "Software systems for complex tasks - such as controlling manufacturing processes in real-time; or writing radiological case reports within a clinical workflow – are becoming increasingly sophisticated and consist of a large number of data, model, algorithm, and system elements and modules. Traditional benchmark/leaderboard-driven bespoke approaches in the Machine Learning community are not suited to meet the highly demanding industrial standards beyond algorithmic performance, such as cost-effectiveness, safety, scalability, and automatability, typically expected in production systems. In this talk, I discuss some technical issues toward addressing these challenges: 1) a theoretical framework for trustworthy and panoramic learning with all experiences; 2) optimization methods to best the effort for learning under such a principled framework; 3) compositional strategies for building production-grade ML programs from standard parts. I will present our recent work toward developing a standard model for Learning that unifies different machine learning paradigms and algorithms, then a Bayesian blackbox optimization approach to Meta Learning in the space of hyperparameters, model architectures, and system configurations, and finally principles and designs of standardized software Legos that facilitate cost-effective building, training, and tunning of practical ML pipelines and systems."
-}
-
-@article{kdddld2021,
- abbr={KDD DLD},
- title = "It is time for deep learning to understand its expense bills",
- year = 2021,
- journal = {KDD Deep Learning Day},
- talk = "https://youtu.be/1ziZcHRqtNU",
- abstract = {In the past several years, deep learning has dominated both academic and industrial R&D over a wide range of applications, with two remarkable trends: 1) developing and training ever larger "all-purpose" monster models over all data possibly available, with a astounding 10,000x parameter number increase in recent 3 years; 2) developing and assembling end-to-end "white-boxes" deployments with ever larger number of component sub-models that need to be highly customized and interoperative. Progresses made to the leaderboards or featured in news headlines are highlighting metrics such as saliency of content production, accuracy on labeling, or speed of convergence, but a number of key challenges impacting the cost effectiveness of such results, and eventually the sustainability of current R&D efforts in DL, are not receiving enough attention: 1) For large models, how many lines of code outside of the DL model are need to parallelize the computing over a computer cluster? (2) Which/How many hardware resources to use to train and deploy the model? (3) How to tune the model, the code, and the system to achieve optimum performance? (4) Can we automate composition, parallelization, tuning, and resource sharing between many users and jobs? In this talk, I will discuss these issues as a core focus in SysML research, and I will present some preliminary results on how to build standardizable, adaptive, and automatable system support for DL based on first principles (when available) underlying DL design and implementation.}
-}
-
-@article{acl2021-workshop,
- abbr={ACL Meta-NLP},
- title = "Learning-to-learn through Model-based Optimization: HPO, NAS, and Distributed Systems",
- year = 2021,
- journal = {ACL 2021 workshop on Meta Learning and Its Applications to Natural Language Processing},
- talk = "https://youtu.be/j66_e7yVoAs",
- abstract = {In recent years we have seen rapid progress in developing modern NLP applications, by either building omni-purpose systems via training massive language models such as GPT-3 on big data, or building industrial solutions for specific real-world use cases via composition from pre-made modules. In both cases, a bottleneck developers often face is the effort required to determine the best way to train the model: such as how to tune the optimal configuration of hyper-parameters of the model(s), big or small, single or multiple; how to choose the best structure of a single large network or a pipeline of multiple model modules; or even how to dynamically pick the best learning rate and gradient-update transmission/synchronization scheme to achieve best “Goodput” of training on a cluster. This is a special area in meta-learning that concerns the question of “learning to learn”. However, many existing methods remain rather primitive, including random search, simple line or grid (or hyper-grid) search, and genetic algorithms, which suffer many limitations such as optimality, efficiency, scalability, adaptability, and ability to leverage domain knowledge.
-In this talk, we present a learning-to-learn methodology based on model-based optimization (MBO), which leverages machine learning models which take actions to gather information and provide recommendations to efficiently improve performance. This exhibits several advantages over existing alternatives: 1) provides adaptive/elastic algorithms that improve performance online; 2) we can incorporate domain knowledge into these models for improved recommendations; 3) can easily facilitate more-data-efficient automatic learning-to-learn, or Auto-ML. We show applications of Auto-ML via MBO in three main tasks: hyper-parameter tuning, neural architecture search, and Goodput optimization in distributed systems. We argue that such applications can improve productivity and performance of NLP systems across the board.}
-}
-
-
-
-@article{icmlml4data2021,
- abbr={ICML ML4Data},
- title = "A Data-Centric View for Composable Natural Language Processing",
- year = 2021,
- journal = {ICML2021 ML4data Workshop},
- talk = "https://youtu.be/JV2y4cT56YE",
- abstract = {Empirical natural language processing (NLP) systems in application domains such as healthcare, finance, and education involve frequent manipulation of data and interoperation among multiple components, ranging from data ingestion, text retrieval, analysis, generation, and even human interactions like visualization and annotation. The diverse nature of the components in such complex systems imposes challenges to create standardized, robust and reusable components.
-In this talk, we present a data centric view of NLP operation and tooling, which bridges different style of software libraries, different user personas, and over additional infrastructures such as those for visualization and distributed training. We propose a highly universal data representation called DataPack, which builds on a flexible type-ontology that is morphable and extendable to subsume any commonly used data formats in all known (and hopefully, future) NLP tasks, yet remains invariant as a software data structure that can be passed across any NLP building blocks. Based on this abstraction, we develop Forte, a Data-Centric Framework for Composable NLP Workflows, with rich in-house processors, standardized 3rd-party API wrappers, and operation logics implemented at the right level of abstraction to facilitate rapid composition of sophisticated NLP solutions with heterogeneous components.
-By defining and leveraging appropriate abstractions of NLP data, Forte aims bridge silos and divergent efforts in NLP tool development, bring good software engineering practices into NLP development, with the goal to help NLP practitioners to build robust NLP systems more efficiently.}
-}
-
diff --git a/_config.yml b/_config.yml
deleted file mode 100644
index 42c6980e..00000000
--- a/_config.yml
+++ /dev/null
@@ -1,242 +0,0 @@
-# -----------------------------------------------------------------------------
-# Site settings
-# -----------------------------------------------------------------------------
-
-title: blank # the website title (if blank, full name will be used instead)
-first_name: Sailing
-middle_name:
-last_name: Lab
-email: you@example.com
-description: > # the ">" symbol means to ignore newlines until "footer_text:"
- A simple, whitespace theme for academics. Based on [*folio](https://github.com/bogoli/-folio) design.
-footer_text: >
- Powered by Jekyll with al-folio theme.
- Hosted by GitHub Pages.
- Photos from Unsplash.
-
-icon: # the emoji used as the favicon
-url: https://sailing-lab.github.io # the base hostname & protocol for your site
-baseurl: # the subpath of your site, e.g. /blog/
-last_updated: false # set to true if you want to display last updated in the footer
-impressum_path: # set to path to include impressum link in the footer, use the same path as permalink in a page, helps to conform with EU GDPR
-
-# -----------------------------------------------------------------------------
-# RSS Feed
-# -----------------------------------------------------------------------------
-# will use title and url fields
-# Take a look to https://github.com/jekyll/jekyll-feed for more customization
-
-# -----------------------------------------------------------------------------
-# Layout
-# -----------------------------------------------------------------------------
-
-navbar_fixed: true
-footer_fixed: true
-
-# Dimensions
-max_width: 800px
-
-# TODO: add layout settings (single page vs. multi-page)
-
-# -----------------------------------------------------------------------------
-# Open Graph
-# -----------------------------------------------------------------------------
-# Display links to the page with a preview object on social media.
-serve_og_meta: false # Include Open Graph meta tags in the HTML head
-og_image: # The site-wide (default for all links) Open Graph preview image
-
-# -----------------------------------------------------------------------------
-# Social integration
-# -----------------------------------------------------------------------------
-
-github_username: # your GitHub user name
-gitlab_username: # your GitLab user name
-twitter_username: # your Twitter handle
-linkedin_username: # your LinkedIn user name
-scholar_userid: # your Google Scholar ID
-orcid_id: # your ORCID ID
-medium_username: # your Medium username
-quora_username: # your Quora username
-publons_id: # your ID on Publons
-research_gate_profile: # your profile on ResearchGate
-blogger_url: # your blogger URL
-work_url: # work page URL
-keybase_username: # your keybase user name
-wikidata_id: # your wikidata id
-dblp_url: # your DBLP profile url
-stackoverflow_id: #your stackoverflow id
-
-rss_icon: true
-
-contact_note: >
- You can even add a little note about which of these is the best way to reach you.
-
-google_analytics: UA-XXXXXXXXX # out your google-analytics code
-panelbear_analytics: XXXXXXXXX # panelbear analytics site ID
-
-# # -----------------------------------------------------------------------------
-# # Blog
-# # -----------------------------------------------------------------------------
-
-# blog_name: al-folio # your blog must have a name for it to show up in the nav bar
-# blog_description: a simple whitespace theme for academics
-# permalink: /blog/:year/:title/
-
-# # Pagination
-# pagination:
-# enabled: true
-
-# # Comments
-# disqus_shortname: al-folio # put your disqus shortname
-# # https://help.disqus.com/en/articles/1717111-what-s-a-shortname
-
-# -----------------------------------------------------------------------------
-# Collections
-# -----------------------------------------------------------------------------
-
-collections:
- news:
- defaults:
- layout: post
- output: true
- permalink: /news/:path/
- projects:
- output: true
- permalink: /projects/:path/
-
-news_limit: 30
-
-# -----------------------------------------------------------------------------
-# Jekyll settings
-# -----------------------------------------------------------------------------
-
-# Markdown and syntax highlight
-markdown: kramdown
-highlighter: rouge
-highlight_theme: github # https://github.com/jwarby/jekyll-pygments-themes
-kramdown:
- input: GFM
- syntax_highlighter_opts:
- css_class: 'highlight'
- span:
- line_numbers: false
- block:
- line_numbers: false
- start_line: 1
-
-# Includes & excludes
-include: ['_pages']
-exclude:
- - bin
- - Gemfile
- - Gemfile.lock
- - vendor
-keep_files:
- - CNAME
- - .nojekyll
- - .git
-
-# Plug-ins
-plugins:
- - jekyll/scholar
- - jekyll-diagrams
- - jekyll-email-protect
- - jekyll-feed
- - jekyll-github-metadata
- - jekyll-paginate-v2
- - jekyll-sitemap
- - jekyll-target-blank
- - jekyll-twitter-plugin
- - jemoji
-
-# Extras
-github: [metadata]
-
-# -----------------------------------------------------------------------------
-# Jekyll Scholar
-# -----------------------------------------------------------------------------
-
-scholar:
-
- last_name: Einstein
- first_name: [Albert, A.]
-
- style: apa
- locale: en
-
- source: /_bibliography/
- bibliography: papers.bib
- bibliography_template: bib
- # Note: if you have latex math in your bibtex, the latex filter
- # preprocessing may conflict with MathJAX if the latter is enabled.
- # See https://github.com/alshedivat/al-folio/issues/357.
- bibtex_filters: [latex, smallcaps, superscript]
-
- replace_strings: true
- join_strings: true
-
- details_dir: bibliography
- details_layout: bibtex.html
- details_link: Details
-
- query: "@*"
-
-# -----------------------------------------------------------------------------
-# Jekyll Diagrams
-# -----------------------------------------------------------------------------
-
-jekyll-diagrams:
- # configuration, see https://github.com/zhustec/jekyll-diagrams.
- # feel free to comment out this section if not using jekyll diagrams.
-
-# -----------------------------------------------------------------------------
-# Optional Features
-# -----------------------------------------------------------------------------
-
-enable_google_analytics: false # enables google analytics
-enable_panelbear_analytics: false # enables panelbear analytics
-enable_mansory: true # enables automatic project cards arangement
-enable_math: true # enables math typesetting (uses MathJax)
-enable_tooltips: false # enables automatic tooltip links generated
- # for each section titles on pages and posts
-enable_darkmode: true # enables switching between light/dark modes
-enable_navbar_social: false # enables displaying social links in the
- # navbar on the about page
-enable_project_categories: true # enables categorization of projects into
- # multiple categories
-enable_medium_zoom: true # enables image zoom feature (as on medium.com)
-
-# -----------------------------------------------------------------------------
-# Library versions
-# -----------------------------------------------------------------------------
-
-academicons:
- version: "1.9.0"
- integrity: "sha512-W4yqoT1+8NLkinBLBZko+dFB2ZbHsYLDdr50VElllRcNt2Q4/GSs6u71UHKxB7S6JEMCp5Ve4xjh3eGQl/HRvg=="
-bootstrap:
- version: "4.5.2"
- integrity:
- css: "sha512-MoRNloxbStBcD8z3M/2BmnT+rg4IsMxPkXaGh2zD6LGNNFE80W3onsAhRcMAMrSoyWL9xD7Ert0men7vR8LUZg=="
- js: "sha512-M5KW3ztuIICmVIhjSqXe01oV2bpe248gOxqmlcYrEzAvws7Pw3z6BK0iGbrwvdrUQUhi3eXgtxp5I8PDo9YfjQ=="
-fontawesome:
- version: "5.14.0"
- integrity: "sha512-1PKOgIY59xJ8Co8+NE6FZ+LOAZKjy+KY8iq0G4B3CyeY6wYHN3yt9PW0XpSriVlkMXe40PTKnXrLnZ9+fkDaog=="
-jquery:
- version: "3.5.1"
- integrity: "sha512-bLT0Qm9VnAYZDflyKcBaQ2gg0hSYNQrJ8RilYldYQ1FxQYoCLtUjuuRuZo+fjqhx/qtq/1itJ0C2ejDxltZVFg=="
-mathjax:
- version: "3.2.0"
-mansory:
- version: "4.2.2"
- integrity: "sha256-Nn1q/fx0H7SNLZMQ5Hw5JLaTRZp0yILA/FRexe19VdI="
-mdb:
- version: "4.19.1"
- integrity:
- css: "sha512-RO38pBRxYH3SoOprtPTD86JFOclM51/XTIdEPh5j8sj4tp8jmQIx26twG52UaLi//hQldfrh7e51WzP9wuP32Q=="
- js: "sha512-Mug9KHKmroQFMLm93zGrjhibM2z2Obg9l6qFG2qKjXEXkMp/VDkI4uju9m4QKPjWSwQ6O2qzZEnJDEeCw0Blcw=="
-popper:
- version: "2.4.4"
- integrity: "sha512-eUQ9hGdLjBjY3F41CScH3UX+4JDSI9zXeroz7hJ+RteoCaY+GP/LDoM8AO+Pt+DRFw3nXqsjh9Zsts8hnYv8/A=="
-medium_zoom:
- version: "1.0.6"
- integrity: "sha256-EdPgYcPk/IIrw7FYeuJQexva49pVRZNmt3LculEr7zM="
diff --git a/_data/coauthors.yml b/_data/coauthors.yml
deleted file mode 100644
index 0c1c954a..00000000
--- a/_data/coauthors.yml
+++ /dev/null
@@ -1,18 +0,0 @@
-"Adams":
- - firstname: ["Edwin", "E.", "E. P.", "Edwin Plimpton"]
- url: https://en.wikipedia.org/wiki/Edwin_Plimpton_Adams
-
-"Podolsky":
- - firstname: ["Boris", "B.", "B. Y.", "Boris Yakovlevich"]
- url: https://en.wikipedia.org/wiki/Boris_Podolsky
-
-"Rosen":
- - firstname: ["Nathan", "N."]
- url: https://en.wikipedia.org/wiki/Nathan_Rosen
-
-"Bach":
- - firstname: ["Johann Sebastian", "J. S."]
- url: https://en.wikipedia.org/wiki/Johann_Sebastian_Bach
-
- - firstname: ["Carl Philipp Emanuel", "C. P. E."]
- url: https://en.wikipedia.org/wiki/Carl_Philipp_Emanuel_Bach
diff --git a/_includes/footer.html b/_includes/footer.html
deleted file mode 100644
index 2e345218..00000000
--- a/_includes/footer.html
+++ /dev/null
@@ -1,27 +0,0 @@
-{% if site.footer_fixed %}
-
-{% else %}
-
-{% endif %}
diff --git a/_includes/head.html b/_includes/head.html
deleted file mode 100644
index 3b1f1f33..00000000
--- a/_includes/head.html
+++ /dev/null
@@ -1,74 +0,0 @@
-
-
-
-
-
-{% if site.title == "blank" %}
- {{ site.first_name }} {{ site.middle_name }} {{ site.last_name }}
-{% else %}
- {{ site.title }}
-{% endif %}
-{% if page.title != "blank" and page.url != "/" %}
- | {{ page.title }}
-{% endif %}
-
-
-
-
-{% if site.serve_og_meta %}
-
-
-
-
-
-
-{% endif %}
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-{% if site.icon != empty %}
-
-{% endif %}
-
-
-
-
-{% include scripts/jquery.html %}
-
-
-{% if site.enable_darkmode %}
-
-
-{% endif %}
-
-{% if site.enable_google_analytics %}
-
-
-
-{% endif %}
-
-{% if site.enable_panelbear_analytics %}
-
-
-
-{% endif %}
diff --git a/_includes/header.html b/_includes/header.html
deleted file mode 100644
index 31e1fd2a..00000000
--- a/_includes/header.html
+++ /dev/null
@@ -1,92 +0,0 @@
-
-
-
-
-
-
diff --git a/_includes/news.html b/_includes/news.html
deleted file mode 100644
index e1fc80bc..00000000
--- a/_includes/news.html
+++ /dev/null
@@ -1,24 +0,0 @@
-
-
news
- {% if site.news %}
-
-
- {% assign news = site.news | reverse %}
- {% for item in news limit: site.news_limit %}
-
diff --git a/_news/group_website.md b/_news/group_website.md
deleted file mode 100644
index b5750534..00000000
--- a/_news/group_website.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: Oct 26, 2021
-inline: true
----
-
-The (renewed) group website is now live!
diff --git a/_news/july_25_2021_eric_on_wired.md b/_news/july_25_2021_eric_on_wired.md
deleted file mode 100644
index ce1eb230..00000000
--- a/_news/july_25_2021_eric_on_wired.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: July 25, 2021
-inline: true
----
-
-Eric is featured on Wired!
diff --git a/_news/new_members_202110.md b/_news/new_members_202110.md
deleted file mode 100644
index 31048e81..00000000
--- a/_news/new_members_202110.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: Oct 1, 2021
-inline: true
----
-
-Welcome new members to the [team](/people).
diff --git a/_news/oct_8_2021_cmu_pf.md b/_news/oct_8_2021_cmu_pf.md
deleted file mode 100644
index 7b1c2446..00000000
--- a/_news/oct_8_2021_cmu_pf.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: Oct 8, 2021
-inline: true
----
-
-Congratulations to Sang for the CMU Presidential Fellowship!
diff --git a/_news/osdi_award.md b/_news/osdi_award.md
deleted file mode 100644
index 23ce83b9..00000000
--- a/_news/osdi_award.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: Jul 27, 2021
-inline: true
----
-
-Congratulations to the SAILING members for winning this year's OSDI 2021 [Best Paper Award](https://www.usenix.org/conference/osdi21/presentation/qiao)!
diff --git a/_news/sept_1_2021_dissertations.md b/_news/sept_1_2021_dissertations.md
deleted file mode 100644
index dd281f3c..00000000
--- a/_news/sept_1_2021_dissertations.md
+++ /dev/null
@@ -1,7 +0,0 @@
----
-layout: post
-date: Sept 1, 2021
-inline: true
----
-
-Congratulations to Maruan, Lisa, and Aurick for successfully defending their dissertations.
diff --git a/_pages/about.md b/_pages/about.md
deleted file mode 100644
index c54c6842..00000000
--- a/_pages/about.md
+++ /dev/null
@@ -1,20 +0,0 @@
----
-layout: about
-title: about
-permalink: /
-description:
-
-profile:
- align: right
- image: sailing-locations.png
- address:
-
-news: true # includes a list of news items
-selected_papers: true # includes a list of papers marked as "selected={true}"
-social: true # includes social icons at the bottom of the page
----
-
-Welcome to the SAILING (Statistical Artificial Intelligence & Integrative Genomics) Lab! SAILING is a research laboratory created in 2004, and is headed by Professor Eric Xing. We are primarily based at Carnegie Mellon University and Mohamed bin Zayed University of Artificial Intelligence.
-
-**Research synopsis**: Our principal research interests lie in the development of machine learning and statistical methodology, and large-scale computational system and architecture, for solving problems involving automated learning, reasoning, and decision-making in high-dimensional, multimodal, and dynamic possible worlds in artificial, biological, and social systems.
-
diff --git a/_pages/community.md b/_pages/community.md
deleted file mode 100644
index cdefc565..00000000
--- a/_pages/community.md
+++ /dev/null
@@ -1,10 +0,0 @@
----
-layout: page
-title: community
-permalink: /community/
-description: Startups and academic labs from SAILING members and alumni.
-nav: true
----
-
-
-
diff --git a/_pages/open-source.md b/_pages/open-source.md
deleted file mode 100644
index 8da1e6ac..00000000
--- a/_pages/open-source.md
+++ /dev/null
@@ -1,60 +0,0 @@
----
-layout: page
-title: open-source
-permalink: /open-source/
-description: On June 11th, 2020, we launched the CASL (Composible, Automatic, and Scable ML) open source consortium that brings our research and development at Petuum Inc. and CMU Sailing Lab on Distributed ML (e.g., AutoDist, AdaptDL), Automated ML (e.g., Dragonfly, ProBO), and Composable ML (e.g., Texar, Forte) implemented across PyTorch and TensorFlow under a unified umbrella for a Production and Industrial AI Platform.
-nav: true
-display_categories: [project]
-horizontal: false
----
-
-SAILING Members: If you want to add your open-source project to this page, please refer to the instruction.
-
-
- {% if site.enable_project_categories and page.display_categories %}
-
- {% for category in page.display_categories %}
-
Many of the problems in artificial intelligence, statistics, computer systems, computer vision, natural language processing, and computational biology, among many other fields, can be viewed as the search for a coherent global conclusion from local information. The probabilistic graphical models framework provides an unified view for this wide range of problems, enables efficient inference, decision-making and learning in problems with a very large number of attributes and huge datasets. This graduate-level course will provide you with a strong foundation for both applying graphical models to complex problems and for addressing core research topics in graphical models.
Machine Learning is concerned with computer programs that automatically improve their performance through experience (e.g., programs that learn to recognize human faces, recommend music and movies, and drive autonomous robots). This course covers the theory and practical algorithms for machine learning from a variety of perspectives. We cover topics such as Linear Regression, SVMs, Neural Networks, Graphical Models, Clustering, etc. Programming assignments include hands-on experiments with various learning algorithms. This course is designed to give a PhD-level student a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who do research in machine learning.
-
-
-
-
-
talks
-{% bibliography -f talks %}
-
-
-
diff --git a/_posts/2015-03-15-formatting-and-links.md b/_posts/2015-03-15-formatting-and-links.md
deleted file mode 100644
index 8edeba65..00000000
--- a/_posts/2015-03-15-formatting-and-links.md
+++ /dev/null
@@ -1,28 +0,0 @@
----
-layout: post
-title: a post with formatting and links
-date: 2015-03-15 16:40:16
-description: march & april, looking forward to summer
----
-Jean shorts raw denim Vice normcore, art party High Life PBR skateboard stumptown vinyl kitsch. Four loko meh 8-bit, tousled banh mi tilde forage Schlitz dreamcatcher twee 3 wolf moon. Chambray asymmetrical paleo salvia, sartorial umami four loko master cleanse drinking vinegar brunch. Pinterest DIY authentic Schlitz, hoodie Intelligentsia butcher trust fund brunch shabby chic Kickstarter forage flexitarian. Direct trade cold-pressed meggings stumptown plaid, pop-up taxidermy. Hoodie XOXO fingerstache scenester Echo Park. Plaid ugh Wes Anderson, freegan pug selvage fanny pack leggings pickled food truck DIY irony Banksy.
-
-#### Hipster list
-
-
brunch
-
fixie
-
raybans
-
messenger bag
-
-
-Hoodie Thundercats retro, tote bag 8-bit Godard craft beer gastropub. Truffaut Tumblr taxidermy, raw denim Kickstarter sartorial dreamcatcher. Quinoa chambray slow-carb salvia readymade, bicycle rights 90's yr typewriter selfies letterpress cardigan vegan.
-
-
-
-Pug heirloom High Life vinyl swag, single-origin coffee four dollar toast taxidermy reprehenderit fap distillery master cleanse locavore. Est anim sapiente leggings Brooklyn ea. Thundercats locavore excepteur veniam eiusmod. Raw denim Truffaut Schlitz, migas sapiente Portland VHS twee Bushwick Marfa typewriter retro id keytar.
-
-
- We do not grow absolutely, chronologically. We grow sometimes in one dimension, and not in another, unevenly. We grow partially. We are relative. We are mature in one realm, childish in another.
- —Anais Nin
-
-
-Fap aliqua qui, scenester pug Echo Park polaroid irony shabby chic ex cardigan church-key Odd Future accusamus. Blog stumptown sartorial squid, gastropub duis aesthetic Truffaut vero. Pinterest tilde twee, odio mumblecore jean shorts lumbersexual.
diff --git a/_posts/2015-05-15-images.md b/_posts/2015-05-15-images.md
deleted file mode 100644
index 63479188..00000000
--- a/_posts/2015-05-15-images.md
+++ /dev/null
@@ -1,45 +0,0 @@
----
-layout: post
-title: a post with images
-date: 2015-05-15 21:01:00
-description: this is what included images could look like
----
-This is an example post with image galleries.
-
-
-
-
-
-
-
-
-
-
- A simple, elegant caption looks good between image rows, after each row, or doesn't have to be there at all.
-
-
-Images can be made zoomable.
-Simply add `data-zoomable` to `` tags that you want to make zoomable.
-
-
-
-
-
-
-
-
-
-
-The rest of the images in this post are all zoomable, arranged into different mini-galleries.
-
-
-
-
-
-
-
-
-
-
-
-
diff --git a/_posts/2015-07-15-code.md b/_posts/2015-07-15-code.md
deleted file mode 100644
index 2dc73ce6..00000000
--- a/_posts/2015-07-15-code.md
+++ /dev/null
@@ -1,39 +0,0 @@
----
-layout: post
-title: a post with code
-date: 2015-07-15 15:09:00
-description: an example of a blog post with some code
----
-This theme implements a built-in Jekyll feature, the use of Rouge, for syntax highlighting.
-It supports more than 100 languages.
-This example is in C++.
-All you have to do is wrap your code in a liquid tag:
-
-{% raw %}
-{% highlight c++ linenos %} code code code {% endhighlight %}
-{% endraw %}
-
-The keyword `linenos` triggers display of line numbers.
-Produces something like this:
-
-{% highlight c++ linenos %}
-
-int main(int argc, char const \*argv[])
-{
- string myString;
-
- cout << "input a string: ";
- getline(cin, myString);
- int length = myString.length();
-
- char charArray = new char * [length];
-
- charArray = myString;
- for(int i = 0; i < length; ++i){
- cout << charArray[i] << " ";
- }
-
- return 0;
-}
-
-{% endhighlight %}
diff --git a/_posts/2015-10-20-comments.md b/_posts/2015-10-20-comments.md
deleted file mode 100644
index 55b900f9..00000000
--- a/_posts/2015-10-20-comments.md
+++ /dev/null
@@ -1,8 +0,0 @@
----
-layout: post
-title: a post with comments
-date: 2015-10-20 11:59:00-0400
-description: an example of a blog post with comments
-comments: true
----
-This post shows how to add DISQUS comments.
diff --git a/_posts/2015-10-20-math.md b/_posts/2015-10-20-math.md
deleted file mode 100644
index c7c2fa2a..00000000
--- a/_posts/2015-10-20-math.md
+++ /dev/null
@@ -1,25 +0,0 @@
----
-layout: post
-title: a post with math
-date: 2015-10-20 11:12:00-0400
-description: an example of a blog post with some math
----
-This theme supports rendering beautiful math in inline and display modes using [MathJax 3](https://www.mathjax.org/) engine. You just need to surround your math expression with `$$`, like `$$ E = mc^2 $$`. If you leave it inside a paragraph, it will produce an inline expression, just like $$ E = mc^2 $$.
-
-To use display mode, again surround your expression with `$$` and place it as a separate paragraph. Here is an example:
-
-$$
-\sum_{k=1}^\infty |\langle x, e_k \rangle|^2 \leq \|x\|^2
-$$
-
-You can also use `\begin{equation}...\end{equation}` instead of `$$` for display mode math.
-MathJax will automatically number equations:
-
-\begin{equation}
-\label{eq:caushy-shwarz}
-\left( \sum_{k=1}^n a_k b_k \right)^2 \leq \left( \sum_{k=1}^n a_k^2 \right) \left( \sum_{k=1}^n b_k^2 \right)
-\end{equation}
-
-and by adding `\label{...}` inside the equation environment, we can now refer to the equation using `\eqref`.
-
-Note that MathJax 3 is [a major re-write of MathJax](https://docs.mathjax.org/en/latest/upgrading/whats-new-3.0.html) that brought a significant improvement to the loading and rendering speed, which is now [on par with KaTeX](http://www.intmath.com/cg5/katex-mathjax-comparison.php).
diff --git a/_posts/2018-12-22-distill.md b/_posts/2018-12-22-distill.md
deleted file mode 100644
index bc19df78..00000000
--- a/_posts/2018-12-22-distill.md
+++ /dev/null
@@ -1,275 +0,0 @@
----
-layout: distill
-title: a distill-style blog post
-description: an example of a distill-style blog post and main elements
-date: 2021-05-22
-
-authors:
- - name: Albert Einstein
- url: "https://en.wikipedia.org/wiki/Albert_Einstein"
- affiliations:
- name: IAS, Princeton
- - name: Boris Podolsky
- url: "https://en.wikipedia.org/wiki/Boris_Podolsky"
- affiliations:
- name: IAS, Princeton
- - name: Nathan Rosen
- url: "https://en.wikipedia.org/wiki/Nathan_Rosen"
- affiliations:
- name: IAS, Princeton
-
-bibliography: 2018-12-22-distill.bib
-
-# Below is an example of injecting additional post-specific styles.
-# If you use this post as a template, delete this _styles block.
-_styles: >
- .fake-img {
- background: #bbb;
- border: 1px solid rgba(0, 0, 0, 0.1);
- box-shadow: 0 0px 4px rgba(0, 0, 0, 0.1);
- margin-bottom: 12px;
- }
- .fake-img p {
- font-family: monospace;
- color: white;
- text-align: left;
- margin: 12px 0;
- text-align: center;
- font-size: 16px;
- }
-
----
-
-**NOTE:**
-Citations, footnotes, and code blocks do not display correctly in the dark mode since distill does not support the dark mode by default.
-If you are interested in correctly adding dark mode support for distill, please open [a discussion](https://github.com/alshedivat/al-folio/discussions) and let us know.
-
-
-## Equations
-
-This theme supports rendering beautiful math in inline and display modes using [MathJax 3](https://www.mathjax.org/) engine.
-You just need to surround your math expression with `$$`, like `$$ E = mc^2 $$`.
-If you leave it inside a paragraph, it will produce an inline expression, just like $$ E = mc^2 $$.
-
-To use display mode, again surround your expression with `$$` and place it as a separate paragraph.
-Here is an example:
-
-$$
-\left( \sum_{k=1}^n a_k b_k \right)^2 \leq \left( \sum_{k=1}^n a_k^2 \right) \left( \sum_{k=1}^n b_k^2 \right)
-$$
-
-Note that MathJax 3 is [a major re-write of MathJax](https://docs.mathjax.org/en/latest/upgrading/whats-new-3.0.html) that brought a significant improvement to the loading and rendering speed, which is now [on par with KaTeX](http://www.intmath.com/cg5/katex-mathjax-comparison.php).
-
-
-***
-
-## Citations
-
-Citations are then used in the article body with the `` tag.
-The key attribute is a reference to the id provided in the bibliography.
-The key attribute can take multiple ids, separated by commas.
-
-The citation is presented inline like this: (a number that displays more information on hover).
-If you have an appendix, a bibliography is automatically created and populated in it.
-
-Distill chose a numerical inline citation style to improve readability of citation dense articles and because many of the benefits of longer citations are obviated by displaying more information on hover.
-However, we consider it good style to mention author last names if you discuss something at length and it fits into the flow well — the authors are human and it’s nice for them to have the community associate them with their work.
-
-***
-
-## Footnotes
-
-Just wrap the text you would like to show up in a footnote in a `` tag.
-The number of the footnote will be automatically generated.This will become a hoverable footnote.
-
-***
-
-## Code Blocks
-
-Syntax highlighting is provided within `` tags.
-An example of inline code snippets: `let x = 10;`.
-For larger blocks of code, add a `block` attribute:
-
-
- var x = 25;
- function(x) {
- return x * x;
- }
-
-
-**Note:** `` blocks do not look well in the dark mode.
-You can always use the default code-highlight using the `highlight` liquid tag:
-
-{% highlight javascript %}
-var x = 25;
-function(x) {
- return x * x;
-}
-{% endhighlight %}
-
-***
-
-## Layouts
-
-The main text column is referred to as the body.
-It is the assumed layout of any direct descendants of the `d-article` element.
-
-
-
.l-body
-
-
-For images you want to display a little larger, try `.l-page`:
-
-
-
.l-page
-
-
-All of these have an outset variant if you want to poke out from the body text a little bit.
-For instance:
-
-
-
.l-body-outset
-
-
-
-
.l-page-outset
-
-
-Occasionally you’ll want to use the full browser width.
-For this, use `.l-screen`.
-You can also inset the element a little from the edge of the browser by using the inset variant.
-
-
-
.l-screen
-
-
-
.l-screen-inset
-
-
-The final layout is for marginalia, asides, and footnotes.
-It does not interrupt the normal flow of `.l-body` sized text except on mobile screen sizes.
-
-
Startups and academic labs from SAILING members and alumni.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/feed.xml b/feed.xml
new file mode 100644
index 00000000..fbc17627
--- /dev/null
+++ b/feed.xml
@@ -0,0 +1,459 @@
+Jekyll2024-02-10T15:25:36+00:00https://sailing-lab.github.io/feed.xmlblankA simple, whitespace theme for academics. Based on [*folio](https://github.com/bogoli/-folio) design.
+a post with diagrams2021-07-04T17:39:00+00:002021-07-04T17:39:00+00:00https://sailing-lab.github.io/2021/07/04/diagramsThis theme supports generating various diagrams from a text description using jekyll-diagrams plugin.
+Below, we generate a few examples of such diagrams using languages such as mermaid, plantuml, vega-lite, etc.
+
+
Note: different diagram-generation packages require external dependencies to be installed on your machine.
+Also, be mindful of that because of diagram generation the fist time you build your Jekyll website after adding new diagrams will be SLOW.
+For any other details, please refer to jekyll-diagrams README.
+
+
Mermaid
+
+
Install mermaid using node.js package manager npm by running the following command:
+
npm install-g mermaid.cli
+
+
+
The diagram below was generated by the following code:
+
+
{% mermaid %}
+sequenceDiagram
+ participant John
+ participant Alice
+ Alice->>John: Hello John, how are you?
+ John-->>Alice: Great!
+{% endmermaid %}
+
+
+
+
+
]]>a distill-style blog post2021-05-22T00:00:00+00:002021-05-22T00:00:00+00:00https://sailing-lab.github.io/2021/05/22/distillNOTE:
+Citations, footnotes, and code blocks do not display correctly in the dark mode since distill does not support the dark mode by default.
+If you are interested in correctly adding dark mode support for distill, please open a discussion and let us know.
+
+
Equations
+
+
This theme supports rendering beautiful math in inline and display modes using MathJax 3 engine.
+You just need to surround your math expression with $$, like $$ E = mc^2 $$.
+If you leave it inside a paragraph, it will produce an inline expression, just like \(E = mc^2\).
+
+
To use display mode, again surround your expression with $$ and place it as a separate paragraph.
+Here is an example:
Citations are then used in the article body with the <d-cite> tag.
+The key attribute is a reference to the id provided in the bibliography.
+The key attribute can take multiple ids, separated by commas.
+
+
The citation is presented inline like this: (a number that displays more information on hover).
+If you have an appendix, a bibliography is automatically created and populated in it.
+
+
Distill chose a numerical inline citation style to improve readability of citation dense articles and because many of the benefits of longer citations are obviated by displaying more information on hover.
+However, we consider it good style to mention author last names if you discuss something at length and it fits into the flow well — the authors are human and it’s nice for them to have the community associate them with their work.
+
+
+
+
Footnotes
+
+
Just wrap the text you would like to show up in a footnote in a <d-footnote> tag.
+The number of the footnote will be automatically generated.This will become a hoverable footnote.
+
+
+
+
Code Blocks
+
+
Syntax highlighting is provided within <d-code> tags.
+An example of inline code snippets: <d-code language="html">let x = 10;</d-code>.
+For larger blocks of code, add a block attribute:
+
+
+ var x = 25;
+ function(x) {
+ return x * x;
+ }
+
+
+
Note:<d-code> blocks do not look well in the dark mode.
+You can always use the default code-highlight using the highlight liquid tag:
+
+
+
+
+
+
Layouts
+
+
The main text column is referred to as the body.
+It is the assumed layout of any direct descendants of the d-article element.
+
+
+
.l-body
+
+
+
For images you want to display a little larger, try .l-page:
+
+
+
.l-page
+
+
+
All of these have an outset variant if you want to poke out from the body text a little bit.
+For instance:
+
+
+
.l-body-outset
+
+
+
+
.l-page-outset
+
+
+
Occasionally you’ll want to use the full browser width.
+For this, use .l-screen.
+You can also inset the element a little from the edge of the browser by using the inset variant.
+
+
+
.l-screen
+
+
+
.l-screen-inset
+
+
+
The final layout is for marginalia, asides, and footnotes.
+It does not interrupt the normal flow of .l-body sized text except on mobile screen sizes.
+
+
+
.l-gutter
+
+
+
Emphasis, aka italics, with asterisks or underscores.
+
+
Strong emphasis, aka bold, with asterisks or underscores.
+
+
Combined emphasis with asterisks and underscores.
+
+
Strikethrough uses two tildes. Scratch this.
+
+
+
First ordered list item
+
Another item
+⋅⋅* Unordered sub-list.
+
Actual numbers don’t matter, just that it’s a number
+⋅⋅1. Ordered sub-list
+
And another item.
+
+
+
⋅⋅⋅You can have properly indented paragraphs within list items. Notice the blank line above, and the leading spaces (at least one, but we’ll use three here to also align the raw Markdown).
+
+
⋅⋅⋅To have a line break without a paragraph, you will need to use two trailing spaces.⋅⋅
+⋅⋅⋅Note that this line is separate, but within the same paragraph.⋅⋅
+⋅⋅⋅(This is contrary to the typical GFM line break behaviour, where trailing spaces are not required.)
URLs and URLs in angle brackets will automatically get turned into links.
+http://www.example.com or http://www.example.com and sometimes
+example.com (but not on Github, for example).
+
+
Some text to show that the reference links can follow later.
No language indicated, so no syntax highlighting.
+But let's throw in a <b>tag</b>.
+
+
+
Colons can be used to align columns.
+
+
+
+
+
Tables
+
Are
+
Cool
+
+
+
+
+
col 3 is
+
right-aligned
+
$1600
+
+
+
col 2 is
+
centered
+
$12
+
+
+
zebra stripes
+
are neat
+
$1
+
+
+
+
+
There must be at least 3 dashes separating each header cell.
+The outer pipes (|) are optional, and you don’t need to make the
+raw Markdown line up prettily. You can also use inline Markdown.
+
+
+
+
+
Markdown
+
Less
+
Pretty
+
+
+
+
+
Still
+
renders
+
nicely
+
+
+
1
+
2
+
3
+
+
+
+
+
+
Blockquotes are very handy in email to emulate reply text.
+This line is part of the same quote.
+
+
+
Quote break.
+
+
+
This is a very long line that will still be quoted properly when it wraps. Oh boy let’s keep writing to make sure this is long enough to actually wrap for everyone. Oh, you can putMarkdown into a blockquote.
+
+
+
Three or more…
+
+
+
+
Hyphens
+
+
+
+
Asterisks
+
+
+
+
Underscores
+
+
Here’s a line for us to start with.
+
+
This line is separated from the one above by two newlines, so it will be a separate paragraph.
+
+
This line is also a separate paragraph, but…
+This line is only separated by a single newline, so it’s a separate line in the same paragraph.
]]>Albert Einsteina post with github metadata2020-09-28T21:01:00+00:002020-09-28T21:01:00+00:00https://sailing-lab.github.io/2020/09/28/github-metadataA sample blog page that demonstrates the accessing of github meta data.
+
+
What does Github-MetaData do?
+
+
Propagates the site.github namespace with repository metadata
+
Setting site variables :
+
+
site.title
+
site.description
+
site.url
+
site.baseurl
+
+
+
Accessing the metadata - duh.
+
Generating edittable links.
+
+
+
Additional Reading
+
+
If you’re recieving incorrect/missing data, you may need to perform a Github API authentication.
+
Go through this README for more details on the topic.
+
This page highlights all the feilds you can access with github-metadata.
+
+
+
+
Example MetaData
+
+
Host Name :
+
URL :
+
BaseURL :
+
Archived :
+
Contributors :
+
]]>a post with twitter2020-09-28T15:12:00+00:002020-09-28T15:12:00+00:00https://sailing-lab.github.io/2020/09/28/twitterA sample blog page that demonstrates the inclusion of Tweets/Timelines/etc.
+
+
Tweet
+
An example of displaying a tweet:
+
jekyll-twitter-plugin (1.0.0): A Liquid tag plugin for Jekyll that renders Tweets from Twitter API http://t.co/m4EIQPM9h4
]]>a post with comments2015-10-20T15:59:00+00:002015-10-20T15:59:00+00:00https://sailing-lab.github.io/2015/10/20/commentsThis post shows how to add DISQUS comments.]]>a post with math2015-10-20T15:12:00+00:002015-10-20T15:12:00+00:00https://sailing-lab.github.io/2015/10/20/mathThis theme supports rendering beautiful math in inline and display modes using MathJax 3 engine. You just need to surround your math expression with $$, like $$ E = mc^2 $$. If you leave it inside a paragraph, it will produce an inline expression, just like \(E = mc^2\).
+
+
To use display mode, again surround your expression with $$ and place it as a separate paragraph. Here is an example:
]]>a post with code2015-07-15T15:09:00+00:002015-07-15T15:09:00+00:00https://sailing-lab.github.io/2015/07/15/codeThis theme implements a built-in Jekyll feature, the use of Rouge, for syntax highlighting.
+It supports more than 100 languages.
+This example is in C++.
+All you have to do is wrap your code in a liquid tag:
+
+
]]>a post with images2015-05-15T21:01:00+00:002015-05-15T21:01:00+00:00https://sailing-lab.github.io/2015/05/15/imagesThis is an example post with image galleries.
+
+
+
+
+
+
+
+
+
+
+ A simple, elegant caption looks good between image rows, after each row, or doesn't have to be there at all.
+
+
+
Images can be made zoomable.
+Simply add data-zoomable to <img> tags that you want to make zoomable.
+
+
+
+
+
+
+
+
+
+
+
The rest of the images in this post are all zoomable, arranged into different mini-galleries.
+
+
+
+
+
+
+
+
+
+
+
+
]]>a post with formatting and links2015-03-15T16:40:16+00:002015-03-15T16:40:16+00:00https://sailing-lab.github.io/2015/03/15/formatting-and-linksJean shorts raw denim Vice normcore, art party High Life PBR skateboard stumptown vinyl kitsch. Four loko meh 8-bit, tousled banh mi tilde forage Schlitz dreamcatcher twee 3 wolf moon. Chambray asymmetrical paleo salvia, sartorial umami four loko master cleanse drinking vinegar brunch. Pinterest DIY authentic Schlitz, hoodie Intelligentsia butcher trust fund brunch shabby chic Kickstarter forage flexitarian. Direct trade cold-pressed meggings stumptown plaid, pop-up taxidermy. Hoodie XOXO fingerstache scenester Echo Park. Plaid ugh Wes Anderson, freegan pug selvage fanny pack leggings pickled food truck DIY irony Banksy.
+
+
Hipster list
+
+
brunch
+
fixie
+
raybans
+
messenger bag
+
+
+
Hoodie Thundercats retro, tote bag 8-bit Godard craft beer gastropub. Truffaut Tumblr taxidermy, raw denim Kickstarter sartorial dreamcatcher. Quinoa chambray slow-carb salvia readymade, bicycle rights 90’s yr typewriter selfies letterpress cardigan vegan.
+
+
+
+
Pug heirloom High Life vinyl swag, single-origin coffee four dollar toast taxidermy reprehenderit fap distillery master cleanse locavore. Est anim sapiente leggings Brooklyn ea. Thundercats locavore excepteur veniam eiusmod. Raw denim Truffaut Schlitz, migas sapiente Portland VHS twee Bushwick Marfa typewriter retro id keytar.
+
+
+ We do not grow absolutely, chronologically. We grow sometimes in one dimension, and not in another, unevenly. We grow partially. We are relative. We are mature in one realm, childish in another.
+ —Anais Nin
+
+
+
Fap aliqua qui, scenester pug Echo Park polaroid irony shabby chic ex cardigan church-key Odd Future accusamus. Blog stumptown sartorial squid, gastropub duis aesthetic Truffaut vero. Pinterest tilde twee, odio mumblecore jean shorts lumbersexual.
Welcome to the SAILING (Statistical Artificial Intelligence & Integrative Genomics) Lab! SAILING is a research laboratory created in 2004, and is headed by Professor Eric Xing. We are primarily based at Carnegie Mellon University and Mohamed bin Zayed University of Artificial Intelligence.
+
+
Research synopsis: Our principal research interests lie in the development of machine learning and statistical methodology, and large-scale computational system and architecture, for solving problems involving automated learning, reasoning, and decision-making in high-dimensional, multimodal, and dynamic possible worlds in artificial, biological, and social systems.
+
+
+
+
+
+
+
news
+
+
+
+
+
+
+
Oct 26, 2021
+
+
+ The (renewed) group website is now live!
+
+
+
+
+
+
+
Oct 8, 2021
+
+
+ Congratulations to Sang for the CMU Presidential Fellowship!
+
+
+
On June 11th, 2020, we launched the CASL (Composible, Automatic, and Scable ML) open source consortium that brings our research and development at Petuum Inc. and CMU Sailing Lab on Distributed ML (e.g., AutoDist, AdaptDL), Automated ML (e.g., Dragonfly, ProBO), and Composable ML (e.g., Texar, Forte) implemented across PyTorch and TensorFlow under a unified umbrella for a Production and Industrial AI Platform.
+
+
+
+
SAILING Members: If you want to add your open-source project to this page, please refer to the instruction.
(Now) Associate Professor, Instituto Superior Técnico; Senior Researcher, Instituto de Telecomunicações; VP of AI Research, Unbabel in Lisbon, Portugal
Acknowledgements: We are grateful to the following agencies/institutions for their contributions to our research: AFOSR, Alfred P. Sloan Foundation, Commonwealth of Pennsylvania, DARPA, Facebook, Glaxo-Smith-Kline, IBM, Intel, National Geospatial-Intelligence Agency, National Institutes of Health, National Science Foundation, Office of Naval Research.
Many of the problems in artificial intelligence, statistics, computer systems, computer vision, natural language processing, and computational biology, among many other fields, can be viewed as the search for a coherent global conclusion from local information. The probabilistic graphical models framework provides an unified view for this wide range of problems, enables efficient inference, decision-making and learning in problems with a very large number of attributes and huge datasets. This graduate-level course will provide you with a strong foundation for both applying graphical models to complex problems and for addressing core research topics in graphical models.
Machine Learning is concerned with computer programs that automatically improve their performance through experience (e.g., programs that learn to recognize human faces, recommend music and movies, and drive autonomous robots). This course covers the theory and practical algorithms for machine learning from a variety of perspectives. We cover topics such as Linear Regression, SVMs, Neural Networks, Graphical Models, Clustering, etc. Programming assignments include hands-on experiments with various learning algorithms. This course is designed to give a PhD-level student a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who do research in machine learning.
+
+
+
+
+
talks
+
+
+
+
+
+
+ CIAI
+
+
+
+
+
+
+
On the Utility of Gradient Compression in Distributed Training Systems
A rich body of prior work has highlighted the existence of communication bottlenecks in distributed training. To alleviate these bottlenecks, a long line of recent research proposes to use gradient compression methods. In this talk, Dr. Hongyi Wang (CMU) will first evaluate gradient compression methods’ efficacy and compare their scalability with optimized implementations of synchronous data-parallel SGD across more than 200 realistic distributed setups. The observation is that, surprisingly, only in six cases out of 200, do gradient compression methods provide promising speedup. He will then introduce our extensive investigation to identify the root causes of this phenomenon and present a performance model that can be used to identify the benefits of gradient compression for a variety of system setups. Finally, Dr. Hongyi will propose a list of desirable properties (along with two algorithmic instances) that a gradient compression method should satisfy, in order for it to provide significant speedup in real distributed training systems.
+
+
+
+
+
+
+
+
+
+
+
+
+
+ Baidu
+
+
+
+
+
+
+
From Learning, to Meta-Learning, to "Lego-Learning” – theory, system, and applications
Software systems for complex tasks - such as controlling manufacturing processes in real-time; or writing radiological case reports within a clinical workflow – are becoming increasingly sophisticated and consist of a large number of data, model, algorithm, and system elements and modules. Traditional benchmark/leaderboard-driven bespoke approaches in the Machine Learning community are not suited to meet the highly demanding industrial standards beyond algorithmic performance, such as cost-effectiveness, safety, scalability, and automatability, typically expected in production systems. In this talk, I discuss some technical issues toward addressing these challenges: 1) a theoretical framework for trustworthy and panoramic learning with all experiences; 2) optimization methods to best the effort for learning under such a principled framework; 3) compositional strategies for building production-grade ML programs from standard parts. I will present our recent work toward developing a standard model for Learning that unifies different machine learning paradigms and algorithms, then a Bayesian blackbox optimization approach to Meta Learning in the space of hyperparameters, model architectures, and system configurations, and finally principles and designs of standardized software Legos that facilitate cost-effective building, training, and tunning of practical ML pipelines and systems.
+
+
+
+
+
+
+
+
+
+
+
+
+
+ KDD DLD
+
+
+
+
+
+
+
It is time for deep learning to understand its expense bills
In the past several years, deep learning has dominated both academic and industrial R&D over a wide range of applications, with two remarkable trends: 1) developing and training ever larger "all-purpose" monster models over all data possibly available, with a astounding 10,000x parameter number increase in recent 3 years; 2) developing and assembling end-to-end "white-boxes" deployments with ever larger number of component sub-models that need to be highly customized and interoperative. Progresses made to the leaderboards or featured in news headlines are highlighting metrics such as saliency of content production, accuracy on labeling, or speed of convergence, but a number of key challenges impacting the cost effectiveness of such results, and eventually the sustainability of current R&D efforts in DL, are not receiving enough attention: 1) For large models, how many lines of code outside of the DL model are need to parallelize the computing over a computer cluster? (2) Which/How many hardware resources to use to train and deploy the model? (3) How to tune the model, the code, and the system to achieve optimum performance? (4) Can we automate composition, parallelization, tuning, and resource sharing between many users and jobs? In this talk, I will discuss these issues as a core focus in SysML research, and I will present some preliminary results on how to build standardizable, adaptive, and automatable system support for DL based on first principles (when available) underlying DL design and implementation.
+
+
+
+
+
+
+
+
+
+
+
+
+
+ ACL Meta-NLP
+
+
+
+
+
+
+
Learning-to-learn through Model-based Optimization: HPO, NAS, and Distributed Systems
In recent years we have seen rapid progress in developing modern NLP applications, by either building omni-purpose systems via training massive language models such as GPT-3 on big data, or building industrial solutions for specific real-world use cases via composition from pre-made modules. In both cases, a bottleneck developers often face is the effort required to determine the best way to train the model: such as how to tune the optimal configuration of hyper-parameters of the model(s), big or small, single or multiple; how to choose the best structure of a single large network or a pipeline of multiple model modules; or even how to dynamically pick the best learning rate and gradient-update transmission/synchronization scheme to achieve best “Goodput” of training on a cluster. This is a special area in meta-learning that concerns the question of “learning to learn”. However, many existing methods remain rather primitive, including random search, simple line or grid (or hyper-grid) search, and genetic algorithms, which suffer many limitations such as optimality, efficiency, scalability, adaptability, and ability to leverage domain knowledge.
+In this talk, we present a learning-to-learn methodology based on model-based optimization (MBO), which leverages machine learning models which take actions to gather information and provide recommendations to efficiently improve performance. This exhibits several advantages over existing alternatives: 1) provides adaptive/elastic algorithms that improve performance online; 2) we can incorporate domain knowledge into these models for improved recommendations; 3) can easily facilitate more-data-efficient automatic learning-to-learn, or Auto-ML. We show applications of Auto-ML via MBO in three main tasks: hyper-parameter tuning, neural architecture search, and Goodput optimization in distributed systems. We argue that such applications can improve productivity and performance of NLP systems across the board.
+
+
+
+
+
+
+
+
+
+
+
+
+
+ ICML ML4Data
+
+
+
+
+
+
+
A Data-Centric View for Composable Natural Language Processing
Empirical natural language processing (NLP) systems in application domains such as healthcare, finance, and education involve frequent manipulation of data and interoperation among multiple components, ranging from data ingestion, text retrieval, analysis, generation, and even human interactions like visualization and annotation. The diverse nature of the components in such complex systems imposes challenges to create standardized, robust and reusable components.
+In this talk, we present a data centric view of NLP operation and tooling, which bridges different style of software libraries, different user personas, and over additional infrastructures such as those for visualization and distributed training. We propose a highly universal data representation called DataPack, which builds on a flexible type-ontology that is morphable and extendable to subsume any commonly used data formats in all known (and hopefully, future) NLP tasks, yet remains invariant as a software data structure that can be passed across any NLP building blocks. Based on this abstraction, we develop Forte, a Data-Centric Framework for Composable NLP Workflows, with rich in-house processors, standardized 3rd-party API wrappers, and operation logics implemented at the right level of abstraction to facilitate rapid composition of sophisticated NLP solutions with heterogeneous components.
+By defining and leveraging appropriate abstractions of NLP data, Forte aims bridge silos and divergent efforts in NLP tool development, bring good software engineering practices into NLP development, with the goal to help NLP practitioners to build robust NLP systems more efficiently.