Ryan PégoudinTowards Data ScienceBreaking down State-of-the-Art PPO Implementations in JAXAll the tricks and details you wish you knew about Proximal Policy Optimization9 min read·May 1, 2024--1--1
Ryan PégoudinTowards Data ScienceA Gentle Introduction to Deep Reinforcement Learning in JAXSolving the CartPole environment with DQN in under a second10 min read·Nov 21, 2023--2--2
Ryan PégoudinTowards Data ScienceImplementing a Transformer Encoder from Scratch with JAX and Haiku 🤖Understanding the fundamental building blocks of Transformers.12 min read·Nov 7, 2023--3--3
Ryan PégoudinTowards Data ScienceVectorize and Parallelize RL Environments with JAX: Q-learning at the Speed of Light⚡Learn to vectorize a GridWorld environment and train 30 Q-learning agents in parallel on a CPU, at 1.8 million step per seconds!11 min read·Oct 15, 2023--2--2
Ryan PégoudinTowards Data ScienceTemporal-Difference Learning and the importance of exploration: An illustrated guideA comparison of model-free (Q-learning) and model-based (Dyna-Q and Dyna-Q+) TD methods on a dynamic grid world.15 min read·Sep 23, 2023--2--2