-
The following game record is the actual game of the "Two-Headed Dragon".
[KataGo_sample_20221217_0001.txt](https://github.com/lightvector/KataGo/files/10246951/KataGo_sample_20221217_0001.txt)
Whi…
-
alphazero的原文里写的前30步走子设置tau=1,即按照概率随机选取动作。之后设置tau趋于0,再采用概率加上狄利克雷噪声的方式选取动作。
这里的实现好像是tau=1,再加上狄利克雷噪声。
这两种方法有理论上或者直觉上的差异吗?
-
I mentioned this with regard to AlphaZero at talkchess as well. Time management is not some separate thing. It is, and has been a part of the rules of chess for more than 100 years. The point of a …
-
I'm using this gym to experiment with an AlphaZero-like algorithm, starting with a very small board (2x2 or 3x3). In that context it's very easy to have games that result in multi-step cycles, which t…
-
I observed that the policy will be set to noise in "expand_node", but the "update_policy" used during inference (in "process_mini_batch") will directly update the policy to the result of network calcu…
-
Whether it should put more priorities to 2 steps capture mode than 1 step capture mode.
-
Hey, since MuZero is very similar but more general, could you PLEASE do a similar article and repo for that? many applications will do better with a more simple version that doesn't have to scale acr…
-
Hello,
I've been having issues with doing self-play on GPU, and after about a week of experimentation I've realized that it is necessary to use this option if I want to train a model on my custom g…
-
a program like https://projecteuler.net/problem=1 can solve it
-
As you now have a powerful M60 GPU (about 2-3x my GTX 1070), I am wondering what would be the most helpful next step so that I can still contribute.