sukhitashvili/pong
A reimplementation of Andrej Karpathy's repository for an RL self-learning AI agent that learns to play Pong through trial and error, using PyTorch
This project helps machine learning researchers and students understand and experiment with reinforcement learning. It takes raw video frames from the classic game Pong and trains an AI agent to play the game through trial and error. The output is a trained AI model capable of playing Pong effectively, allowing practitioners to observe and analyze the learning process of policy gradient methods.
No commits in the last 6 months.
Use this if you are studying reinforcement learning and want a hands-on example of an agent learning directly from pixel data using policy gradients.
Not ideal if you are looking for a pre-trained, ready-to-use AI for a real-world application or a comprehensive reinforcement learning framework.
Stars
17
Forks
3
Language
Python
License
—
Category
Last pushed
Aug 23, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/sukhitashvili/pong"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Toni-SM/skrl
Modular Reinforcement Learning (RL) library (implemented in PyTorch, JAX, and NVIDIA Warp) with...
facebookresearch/BenchMARL
BenchMARL is a library for benchmarking Multi-Agent Reinforcement Learning (MARL). BenchMARL...
utiasDSL/gym-pybullet-drones
PyBullet Gymnasium environments for single and multi-agent reinforcement learning of quadcopter control
datamllab/rlcard
Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.
gtri/scrimmage
Multi-Agent Robotics Simulator