README.md

metadata

tags:
  - ALE/KungFuMaster-v5
  - A3C
  - ICM
  - reinforcement-learning
  - atari
  - ALE/KungFuMaster-v5
  - KungFuMaster-v5
model-index:
  - name: A3C-ICM-KungFuMaster-v5
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: ALE/KungFuMaster-v5
          type: ALE/KungFuMaster-v5
        metrics:
          - type: mean_reward
            value: 5750.00 +/- 1608.88
            name: mean_reward
            verified: false

A3C with ICM Agent playing ALE/KungFuMaster-v5

This is a trained model of a A3C-ICM agent playing ALE/KungFuMaster-v5 .

Codes

Github repos(Give a star if found useful):

• https://github.com/hishamcse/Advanced-DRL-Renegades-Game-Bots
• https://github.com/hishamcse/DRL-Renegades-Game-Bots
• https://github.com/hishamcse/Robo-Chess

Kaggle Notebook:

• https://www.kaggle.com/code/syedjarullahhisham/drl-extra-personal-adv-drl-a3c-icm-kungfu-master

Hyperparameters

"envid": "ALE/KungFuMaster-v5",
"learningrate": 0.0002,
"tau": 1.0,
"num_local_step": 50,
"num_global_step": 380000,
"gamma": 0.99,
"lambda": 0.1,
"eta": 0.2,
"beta": 0.2,
"max_actions": 500