Michaël Trazzi

Michaël codes at school 42 and is finishing a Master's Degree in AI. He's currently exploring the limits of productivity and the future of AI. You can follow along with Michaël on Medium or Twitter.

📍 Paris

Reinforcement Learning

Meta-Reinforcement Learning

The general trend in machine learning research is to stop fine-tuning models, and instead use a meta-learning algorithm that automatically finds the best architecture and hyperparameters. What about meta-reinforcement learning (meta-RL)? Meta-RL is just meta-learning applied to RL.

Deep Learning

Spinning Up a Pong AI With Deep Reinforcement Learning

Dive into deep reinforcement learning by training a model to play the classic 1970s video game Pong — using Keras, FloydHub, and OpenAI's "Spinning Up."