TJU-DRL-LAB / AI-Optimizer

The next generation deep reinforcement learning tookit

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Is the sampled muzero Implementation complete?

FYQ0919 opened this issue · comments

I compared the sampled muzero code with the muzero general but I didn't find the code about the number of samples and the policy improvement, can you tell me what changes you have made?

Hello, we are working on implementing the sampled muzero but not finished yet. Actually the current version in this repo is exactly general muzero, which is the coding base for developing sampled muzero. However the version we have implemented so far CANNOT reproduce the results in the paper so we don't intend to open source it until we obtain a comparable performance reported in the paper. We'll release the code as soon as we can.

commented

So is the sampled muzero opened now?