OpenDILab Decision AI Engine. The Most Comprehensive Reinforcement Learning Framework B.P.

RL starter files in order to immediately train, visualize and evaluate an agent without writing any line of code

JAX-accelerated Meta-Reinforcement Learning Environments Inspired by XLand and MiniGrid 🏎️

Recurrent and multi-process PyTorch implementation of deep reinforcement Actor-Critic algorithms A2C and PPO

Accelerated minigrid environments with JAX

Python code to implement LLM4Teach, a policy distillation approach for teaching reinforcement learning agents with Large Language Model

An environement builder for hierarchical reasoning research

Site level energy system optimization model toolbox for investigations of EV fleet integration and management strategies

Solving games with reinforcement learning

Learning Visual Embeddings for Reinforcement Learning

Implementation of Offline Reinforcement Learning in Gym Mini-Grid Environment 🔑

An alternate way to view the Minigrid environment using animations and sprites. Intended to be used as a teaching tool for kids to promote more engagement and interest in Reinforcement Learning.

Deep RL agents for the MiniGrid 6 Rooms. Implements DQN, Double DQN, Huber Loss, Soft Updates, PER, and Actor-Critic. Best model solves 90% of episodes with improved training stability.

Project for the course Deep Learning and Applied AI a.y. 2021/2022, Dept. of Computer Science, Prof. Emanuele Rodolà