v3.18.7

   🐞 Bug Fixes

• ai-CNN:
  • Mnist keras example  -  by @sabertazimi (e39d8)
• ai-DL:
  • Variational auto-encoders  -  by @sabertazimi (6975d)
  • Generative adversarial networks  -  by @sabertazimi (78c11)
  • Recurrent neural networks  -  by @sabertazimi (5af5d)
  • Long short-term memory network  -  by @sabertazimi (8771b)
  • Use activation function to fit complex function  -  by @sabertazimi (ce182)
  • Critical point in gradient descent  -  by @sabertazimi (30c18)
  • Saddle point  -  by @sabertazimi (9c346)
  • Small batch vs large batch  -  by @sabertazimi (2f7dd)
  • Learning rate adjustment strategies  -  by @sabertazimi (fdb61)
  • Change loss function for easier optimization  -  by @sabertazimi (5a3a8)
  • Adaptive learning rate  -  by @sabertazimi (6ca40)
  • Convolutional layer  -  by @sabertazimi (1883d)
  • Why deep learning  -  by @sabertazimi (c5b2f)
  • Spatial transformer networks  -  by @sabertazimi (33f6b)
  • Self-attention calculation by matrix multiplication  -  by @sabertazimi (3839d)
  • Alignment vector addition to output  -  by @sabertazimi (ef75f)
  • BERT model  -  by @sabertazimi (e1ff3)
  • Pre-trained models and downstream tasks  -  by @sabertazimi (f4698)
  • Selective synaptic plasticity loss function  -  by @sabertazimi (17c83)
  • BERT adapters  -  by @sabertazimi (3b8e7)
  • In-context learning  -  by @sabertazimi (7b13a)
  • Instruction-tuning  -  by @sabertazimi (dda0c)
  • Architecture of BERT and GPT  -  by @sabertazimi (79240)
• ai-ML:
  • Principal component analysis  -  by @sabertazimi (b6700)
  • Structured learning  -  by @sabertazimi (2a4ab)
  • Structured linear model  -  by @sabertazimi (7ba7c)
  • Reinforcement learning  -  by @sabertazimi (89cc3)
  • Learn from rewards and mistakes  -  by @sabertazimi (a0e35)
  • Underfitting issue  -  by @sabertazimi (c85ac)
  • Extreme example of overfitting  -  by @sabertazimi (fee64)
  • Prevent overfitting  -  by @sabertazimi (20ed8)
  • Normalization  -  by @sabertazimi (ce775)
  • Cycle GAN for unpaired image-to-image translation  -  by @sabertazimi (1f99b)
  • Auto-encoder applications  -  by @sabertazimi (5cfcc)
  • Explainable AI  -  by @sabertazimi (3cb55)
  • Self-supervised learning story  -  by @sabertazimi (5c384)
  • Discounted cumulated reward formula  -  by @sabertazimi (3057b)
  • Actor-critic model for reinforcement learning  -  by @sabertazimi (b65c8)
  • Inverse reinforcement learning  -  by @sabertazimi (4229a)
• ai-NLP:
  • Word embedding  -  by @sabertazimi (36263)
• ai-gen:
  • AR and NAR models  -  by @sabertazimi (362f7)
  • Chain-of-thought prompting  -  by @sabertazimi (0e5ef)
  • Machine prompting  -  by @sabertazimi (db6ec)
  • Prompt engineering guide  -  by @sabertazimi (7140b)
  • Reinforcement learning from human feedback  -  by @sabertazimi (00b45)
  • Diffusion model  -  by @sabertazimi (ee5e5)
  • ChatGPT alignment  -  by @sabertazimi (d0c1d)
  • Labelled data and human feedback  -  by @sabertazimi (ad950)
  • Retrieve-augmented generation  -  by @sabertazimi (fe82f)
  • Recursive re-prompting and revision  -  by @sabertazimi (dd15b)
  • Task decomposition prompting  -  by @sabertazimi (0396e)
  • Retrieval-augmented generation  -  by @sabertazimi (dd87a)
  • LLMs cooperation  -  by @sabertazimi (cf076)
  • Exchange-of-thought  -  by @sabertazimi (1d0b6)
  • Multi-agent debate  -  by @sabertazimi (69c81)
  • Multi-agent collaboration  -  by @sabertazimi (1c6c0)
  • Data for LLMs  -  by @sabertazimi (be4ff)
  • Alignment for LLMs  -  by @sabertazimi (417cb)
  • How to fine-tune LLM  -  by @sabertazimi (3d55c)
  • Reward model as human feedback  -  by @sabertazimi (42c3a)
  • Midjourney cookbook  -  by @sabertazimi (27ce6)
  • AI agents  -  by @sabertazimi (44684)
  • Encoder-decoder generative framework  -  by @sabertazimi (d0804)
  • Speculative decoding  -  by @sabertazimi (90c98)
  • Retrieval-augmented generation example  -  by @sabertazimi (3f17c)
  • RAG workflow diagram  -  by @sabertazimi (0e05b)
  • Human alignment for LLMs  -  by @sabertazimi (d389a)
  • Scaling law and emergent ability  -  by @sabertazimi (27388)
  • RLHF focus on three aspects  -  by @sabertazimi (333f6)
  • Key advantage of self-attention mechanism  -  by @sabertazimi (b4bca)
  • Low-rank adaptation fine-tuning  -  by @sabertazimi (784e7)
  • Good parts of instruction-tuning  -  by @sabertazimi (54c86)
  • Targets of RLHF  -  by @sabertazimi (97c5a)
  • Task planning  -  by @sabertazimi (8d632)
  • LLMs survey  -  by @sabertazimi (972f6)
  • Quality in and quality out  -  by @sabertazimi (27b85)
  • Tree of thoughts example  -  by @sabertazimi (45478)
  • Program-aided prompting  -  by @sabertazimi (32631)
  • ReAct prompting  -  by @sabertazimi (209bc)
  • Basic elements of a prompt  -  by @sabertazimi (1d5a2)
  • 26 design principles for prompt engineering  -  by @sabertazimi (22f36)
  • Prompt compression  -  by @sabertazimi (b204b)
• ai-library:
  • Multi-agent framework  -  by @sabertazimi (23295)
• ai-llm:
  • ChatGPT learning process  -  by @sabertazimi (3664f)
• css-design:
  • Design tokens naming convention  -  by @sabertazimi (34002)
• deps:
  • Update dependencies (non-major)  -  by @renovate[bot] in #533 (21d4e)
  • Update dependencies (non-major)  -  by @renovate[bot] in #537 (569b6)
  • Update dependencies (non-major)  -  by @renovate[bot] in #540 (fc94c)
• react-compiler:
  • useMemoCache hook for React Compiler  -  by @sabertazimi (9a728)
• web-library:
  • Modern execa  -  by @sabertazimi (025e9)
  • Music notation and guitar tablature rendering library  -  by @sabertazimi (e720d)
  • Pglite database  -  by @sabertazimi (335f9)
  • Optimize compiler for React.js  -  by @sabertazimi (a017c)
  • Whiteboard and canvas tool  -  by @sabertazimi (aa881)
  • One time password input  -  by @sabertazimi (bbd76)
  • Human-friendly date picker  -  by @sabertazimi (0bf56)
  • Mouse-tracing glow effect library  -  by @sabertazimi (ea8f0)
  • Schema validation library  -  by @sabertazimi (21835)
  • Geo library  -  by @sabertazimi (45587)
  • Collaborative utils  -  by @sabertazimi (b5774)

   🏎 Performance

• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #531 (0806f)
• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #532 (71f2c)
• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #534 (3d267)
• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #536 (dba8a)
• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #539 (e42c8)
• [ImgBot] optimize images  -  by @imgbot[bot] and ImgBotApp in #541 (2ae4b)

    View changes on GitHub