From 0a95b7c994e364c1c755377a8539da7eab5bc089 Mon Sep 17 00:00:00 2001
From: Gavin Kerrigan <36687848+GavinKerrigan@users.noreply.github.com>
Date: Mon, 22 Apr 2024 13:33:00 -0700
Subject: [PATCH] update invited talks schedule

---
 schedule.md | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/schedule.md b/schedule.md
index 78981d1..bc9edd7 100644
--- a/schedule.md
+++ b/schedule.md
@@ -41,12 +41,12 @@ Opening remarks
 09:00-10:00
 </td>
 <td>
-<a href="{{ "/invited.html" | relative_url }}">Keynote Talk: Aaditya Ramdas (CMU)</a>
+<a href="{{ "/invited.html" | relative_url }}">Keynote Talk: Matt Hoffman (DeepMind)</a>  
 <details>
 <summary>
-Conformal Online Model Aggregation
+TBD
 </summary>
-Conformal prediction equips machine learning models with a reasonable notion of uncertainty quantification without making strong distributional assumptions. It wraps around any black-box prediction model and converts point predictions into set predictions that have a predefined marginal coverage guarantee. However, conformal prediction only works if we fix the underlying machine learning model in advance. A relatively unaddressed issue in conformal prediction is that of model selection and/or aggregation: for a given problem, which of the plethora of prediction methods (random forests, neural nets, regularized linear models, etc.) should we conformalize? This talk presents a new approach towards conformal model aggregation in online settings that is based on combining the prediction sets from several algorithms by voting, where weights on the models are adapted over time based on past performance.
+TBD
 </details>
 </td>
 </tr>
@@ -190,14 +190,13 @@ Mentoring Event | TBD
 09:00-10:00
 </td>
 <td>
-<a href="{{ "/invited.html" | relative_url }}">Keynote Talk: Matt Hoffman (DeepMind)</a>  
+<a href="{{ "/invited.html" | relative_url }}">Keynote Talk: Aaditya Ramdas (CMU)</a>
 <details>
 <summary>
-TBD
+Conformal Online Model Aggregation
 </summary>
-TBD
+Conformal prediction equips machine learning models with a reasonable notion of uncertainty quantification without making strong distributional assumptions. It wraps around any black-box prediction model and converts point predictions into set predictions that have a predefined marginal coverage guarantee. However, conformal prediction only works if we fix the underlying machine learning model in advance. A relatively unaddressed issue in conformal prediction is that of model selection and/or aggregation: for a given problem, which of the plethora of prediction methods (random forests, neural nets, regularized linear models, etc.) should we conformalize? This talk presents a new approach towards conformal model aggregation in online settings that is based on combining the prediction sets from several algorithms by voting, where weights on the models are adapted over time based on past performance.
 </details>
-
 </td>
 </tr>