-
Using checkpoint from results/cartpole/2020-08-31--17-00-26/model.checkpoint
Traceback (most recent call last):
File "muzero.py", line 581, in
muzero.load_model(
File "muzero.py", line 40…
-
Just a totally naive thought. On a simpler basis, would, say, a network with inputs that is intended to take "previous iteration results" (for example, the estimated b/w occupancy of the whole board o…
-
What about to upgrade KataGo for incorporating learned model into the training procedure like in MuZero(https://arxiv.org/pdf/1911.08265.pdf)?
-
Dear Mr. Duvaud,
Upon cloning your repository to ".../Documents/AI", adding swig-4.0.2 (https://sourceforge.net/projects/swig/files/swigwin/swigwin-4.0.2/swigwin-4.0.2.zip/download?use_mirror=pho…
-
Greetings,
Some implementations of MuZero use game symmetries to speed up convergence. Are there plans for implementation in this repo?
-
-
A natural extension of AlphaZero is [MuZero](https://arxiv.org/abs/1911.08265). The pseudocode implementation is very similar to that of AlphaZero, so it would be interesting to extend the AlphaZero i…
-
Mr. Varty,
Would be willing to provide scripts or ideas for self-play, console-IO play, and/or 2D games?
Tom Lever
-
https://github.com/levilelis/h-levin/blob/016c3432c41b4644b806b53ceaad1ca4571d25df/src/search/puct.py#L60
The "-" should probably be a "+".
Maybe follow [MuZero](https://arxiv.org/pdf/1911.08265.p…
-
Is there a reason why on every step of selfplay gave muzero starts to build game tree from scratch?
We use the same networks throughout the game. So game subtree, that we build on second step should …