# vietnh1009/Super-mario-bros-PPO-pytorch

Proximal Policy Optimization (PPO) algorithm for Super Mario Bros

Repository: https://github.com/vietnh1009/Super-mario-bros-PPO-pytorch
Canonical: https://ross.abutalabs.com/products/super-mario-bros-ppo-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: reinforcement-learning, ppo, ppo2, pytorch, gym, python3, python, deep-learning, super-mario-bros, mario, ai, proximal-policy-optimization, openai, openai-gym
Last push: 2021-07-24T11:19:21+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2527, "days_push": 1866, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1299, forks 240 (observed 2026-08-28T04:04:17.340418+00:00)

## What it is
A PyTorch implementation of the Proximal Policy Optimization (PPO) algorithm that trains an agent to play Super Mario Bros via OpenAI Gym. The trained agent completes 31 of 32 levels, and the repo serves as a reference implementation of PPO for game environments.

## Use cases
- train an AI agent to play Super Mario Bros
- learn how PPO works with a concrete PyTorch example
- compare PPO against A3C/A2C on a game environment
- get a starting point for reinforcement learning on OpenAI Gym games
- study policy gradient methods with real training code
- reproduce a high-performing game-playing RL agent

## When to choose
- you want a clear, working PPO reference implementation in PyTorch
- you're learning deep RL and want a complete, well-documented example
- you need a baseline for RL agents on classic NES game environments

## When to avoid
- you need a general-purpose RL library with many algorithms
- you want production-ready or actively maintained RL tooling
- you need multi-agent or non-Atari/non-Mario environment support

## Facets
- artifact type: application
- maturity: maintenance
- function: reinforcement-learning, deep-learning, machine-learning
- domain: reinforcement-learning, artificial-intelligence
- platform: python, windows
- tags: ppo, pytorch, openai-gym, super-mario-bros, game-playing-agent, policy-gradient, game-development, linux, macos

## Member repositories
- vietnh1009/Super-mario-bros-PPO-pytorch (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:17.340418+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:53:28.189761+00:00, confidence not recorded.
  - readme: https://github.com/vietnh1009/Super-mario-bros-PPO-pytorch (fetched 2026-08-28T04:04:17.340418+00:00, sha 362a0995d2c1)
- Data as of 2026-08-30T08:39:29.467469+00:00.