Jannis Blüml

Combining Mixture of Experts and Monte Carlo Tree Search for better chess performance.

A new framework to improve AI agents' learning through modified Atari games.

A new method to define rewards for reinforcement learning agents using language models.