vietnh1009/Super-mario-bros-PPO-pytorch

Proximal Policy Optimization (PPO) algorithm for Super Mario Bros

50
/ 100
Established

Implements a deep reinforcement learning agent using PyTorch with an actor-critic architecture, achieving 31/32 level completion across the full Super Mario Bros NES game by leveraging PPO's clipped objective function for stable policy updates. Integrates with the OpenAI Gym environment wrapper for NES emulation and employs CNN feature extraction from raw pixel inputs paired with separate policy and value networks. Training supports per-level hyperparameter tuning and Docker containerization for reproducible GPU-accelerated training workflows.

1,268 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

1,268

Forks

236

Language

Python

License

MIT

Category

lunar-lander-rl

Last pushed

Jul 24, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/vietnh1009/Super-mario-bros-PPO-pytorch"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.