search

Unity-Technologies / ml-agents

The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.
https://unity.com/products/machine-learning-agents
Other
17.19k stars 4.16k forks source link

Can we use ML algorithms other than PPO in unity mlagents for multi agent like ddpg #2134

Closed Usmaniatech closed 5 years ago

Usmaniatech commented 5 years ago

I want to use multiple algorithms for a multi agent situation including PPO but I can't find any support for other algorithms like DDPG. If it's applicable how can I start working on it, what will be the procedure. How much changes do I have to make in other libraries in PPO.

awjuliani commented 5 years ago

Hi @Usmaniatech

It is possible to extend our training code to include other algorithms. For an example of this, you can see this experimental implementation of Soft Actor Critic: https://github.com/Unity-Technologies/ml-agents/tree/develop-sac/ml-agents/mlagents/trainers/sac/

It is also possible to use ML-Agents environment with any external training code designed to work with the OpenAI gym interface: https://github.com/Unity-Technologies/ml-agents/tree/master/gym-unity

Usmaniatech commented 5 years ago

Thanks @awjuliani

Usmaniatech commented 5 years ago

I am able to use MADDPG in mlagents 0.4 but it's not working with 0.8.1.

harperj commented 5 years ago

Hi all -- this issue has been inactive for some time so I'm going to close it. Feel free to reopen or create a new issue if you have more to discuss.

github-actions[bot] commented 3 years ago

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.