[Up-to-date] A curated list of resources on graph-empowered agents and agent-facilitated graph learning (Graphs Meet Agents).

Non-deterministic, human-like mouse and keyboard automation powered by reinforcement learning.

Deep Reinforcement Learning Texas Hold'em AI / 德州AI/深度强化学习德州扑克AI / 深度強化學習德州撲克AI - CFR, DQN, AlphaZero-style training

Train SLM to use Tools with RL

An interactive browser-based playground to learn reinforcement learning — from bandits to policy gradients. Watch 13 algorithms learn in real time across 4 environments (Bandit, GridWorld, CartPole, Rocket Landing), tune hyperparameters, and follow a built-in 10-chapter course. No backend, no account — just open and learn.

A robust Reinforcement Learning environment for the oink games. Compatible with OpenAI Gym interface.

Reinforcement learning Tetris bot

A smart traffic light management dashboard using SUMO and reinforcement learning to simulate traffic flow, optimize signal timing, and improve vehicle movement at intersections.

Sprite Garden

A robust reinforcement learning framework for voicebot turn-level decision optimization. Utilizes offline RL techniques to learn from historical, noisy, and delayed feedback signals without deploying exploratory policies, ensuring safe and stable conversational improvement.

A high-fidelity OpenEnv simulation training RL agents to autonomously review Pull Requests by navigating repositories and leaving precise, line-level spatial comments on detected vulnerabilities.

A high-density browser-based evolutionary ecosystem simulation combining swarm behaviour, reinforcement-learning-style commander control, environmental pressure, emergent cooperation, resource competition, pathogen dynamics, and research-inspired quality-diversity tracking.

Bitcoin trading agent using Deep Q-Learning and synthetic market scenarios.

A Pong game which you can actually play with a sophisticated RL Agent !!

Automated Text Generation Using English Grammar (AI but AI is a Lie 😂, its Mathematics. AI, LLMs, because these keywords are trending 😂)

🏓 A self-learning Pong agent blending value-based and policy-based RL — playable in your terminal.

Deep Q-Network (DQN) agent trained to play Flappy Bird using PyTorch & Gymnasium — with experience replay, target network, and epsilon-greedy exploration.

A reinforcement learning agent that learns to play Pacman in a custom environment with a visual GUI.