README.md

metadata

tags:
  - ML-Agents-SoccerTwos
  - SoccerTwos
  - ml-agents
  - deep-rl-course
  - ppo
  - reinforcement-learning
  - multi-agent
library_name: ml-agents
pipeline_tag: reinforcement-learning
model-index:
  - name: ppo-SoccerTwos
    results:
      - task:
          name: reinforcement-learning
          type: reinforcement-learning
        dataset:
          name: ML-Agents-SoccerTwos
          type: ML-Agents-SoccerTwos
        metrics:
          - name: mean_reward
            type: mean_reward
            value: 0.5

PPO Agent playing SoccerTwos in Unity ML-Agents

This is a trained multi-agent model playing SoccerTwos using Unity ML-Agents and Self-Play.

• Mean Reward: 0.5 +/- 0.1
• Result (mean - std): 0.40