A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

Single GPU, From Scratch (No RL Library), Efficient, Full Parameter Tuning Implementation of DeepSeek R1-Zero style training.

Train transformer language models with reinforcement learning.

AI researcher - with a single query determine focus areas to investigate, searching the web and scraping content from relevant websites to do research autonomously.

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Stable Diffusion web UI

Thompson Sampling Tutorial

Kalman Filter book using Jupyter Notebook. Focuses on building intuition and experience, not formal proofs. Includes Kalman filters,extended Kalman filters, unscented Kalman filters, particle filte…

Google Research

A benchmark for offline reinforcement learning.

Finetuning any DNN for better embedding on neural search tasks

Reinforcement learning algorithms for MuJoCo tasks

Source code of my personal blog

PyTorch implementation of Advantage Actor Critic (A2C), Proximal Policy Optimization (PPO), Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (ACKT…