In the latest release of the ML-Agents Toolkit (v0.14), we have added a self-play feature that provides the capability to train competitive agents in adversarial games (as in zero-sum games, where one agent's gain is exactly the other agent's loss). In this blog post, we provide an overview of self-play and demonstrate how it enables [...]
The post
Training intelligent adversaries using self-play with ML-Agents appeared first on
Unity Technologies Blog.
Ver artículo completo... (https://blogs.unity3d.com/es/2020/02/28/training-intelligent-adversaries-using-self-play-with-ml-agents/)