https://yugeten.github.io/posts/2025/01/ppogrpo A vision researcher’s guide to some RL stuff: PPO & GRPOYuge (Jimmy) ShiJanuary 31, 2025 “This is a deep dive into Proximal Policy Optimization (PPO), which is one of
What I Read: group relative policy optimization
https://superb-makemake-3a4.notion.site/group-relative-policy-optimization-GRPO-18c41736f0fd806eb39dc35031758885 group relative policy optimization (GRPO)Apoorv NandanJan 31, 2025 “GRPO became popular primarily due to the success of deepseek r1, which used this algorithm to train reasoning capabilities into their
What I Read: VAE
https://www.rehansheikh.com/blog/vae What the F*** is a VAE?Rehan SheikhJanuary 23, 2025 “A disentangled variational autoencoder aims for each latent dimension… to correspond to a single factor of variation in your dataset.”
What I Read: impossible languages
https://www.quantamagazine.org/can-ai-models-show-us-how-people-learn-impossible-languages-point-a-way-20250113/ Can AI Models Show Us How People Learn? Impossible Languages Point a Way.Ben Brubaker1/13/25 11:00 AM “Certain grammatical rules never appear in any known language. By constructing artificial languages
What I Read: transfer learning
https://lunar-joke-35b.notion.site/Transfer-Learning-101-133ba4b6a3fa800e8cede11ee3f1c1cd Transfer Learning 101Himanshu DubeyNov 5, 2024 “Let’s understand Transfer Learning in greater detail.”
What I Read: Model Merging
https://planetbanatt.net/articles/modelmerging.html Model Merging and YouEryk BanattAugust 2024 “Model Merging is a weird and experimental technique which lets you take two models and combine them together to get a new model.”
What I Read: ML, Go
https://eli.thegreenplace.net/2024/gomlx-ml-in-go-without-python/#footnote-reference-2 GoMLX: ML in Go without PythonEli BenderskyNovember 22, 2024 at 07:00 “GoMLX is a relatively new Go package for ML that deserves some attention.”
What I Read: Autoencoders, Interpretability
https://adamkarvonen.github.io/machine_learning/2024/06/11/sae-intuitions.html An Intuitive Explanation of Sparse Autoencoders for LLM InterpretabilityAdam KarvonenJun 11, 2024 “Sparse Autoencoders (SAEs) have recently become popular for interpretability of machine learning models…”