ML-Agents in Unity implements self-play, a mechanism for an agent to play against past versions of itself. This expands the possibilities of reinforcement learning as the environment and rewards are always scaling to the abilities of the agent. I really believe self play to be one of the key ideas in machine learning / artificial intelligence and I hope this video gave you some insight into the idea.
Github-Link: github.com/Sebastian-Schuchma...
I have created a Discord Channel for everybody wanting to learn ML-Agents. It's a place where we can help each other out, ask questions, share ideas, and so on. You can join here: / discord
Support me on Patreon: www.patreon.com/user?u=25285137
Keep in touch (Twitter): / sebastianschuc7
21 авг 2020