-
I am testing a monte carlo tree search type of algorithm, which requires repeatedly making multiple copies of the current environment and playing them out using slightly different policies. Previous u…
-
See [this wiki page](https://github.com/Chicoryn/dream-go/wiki/Reinforcement-Learning---Round-2) for the updated ELO ratings and status of the training procedure.
Changes from the last reinforcemen…
-
First of all thanks for the wunderful program. It is a real pleasure to use.
But I have to say that there is something fundamentally wrong with the algorithm.
The program performs a lot of play-ou…
-
I'm thinking about other improves to main time controls. It is said Golaxy is able to do smart time control and it can sometimes use quite a lot of time for a single move (e.g. 3 minutes).
I wonder…
-
Paper: https://arxiv.org/abs/1802.04697
> Planning problems are among the most important and well-studied problems in artificial intelligence. They are most typically solved by tree search algorith…
-
Hi there!
I've been following this article to learn about Monte Carlo Tree Search, and it's been very useful. Thank you! https://www.baeldung.com/java-monte-carlo-tree-search
Just thought you sh…
-
https://github.com/gcp/leela-zero/blob/94a345f7cfcc47fff8a066e68d95376ccef9cde4/src/UCTNode.cpp#L288
In normal /next we overwrite the fpu value if the node has visits:
```
auto winrate …
-
Using NN for time-management settings and accurate boardstate counting with the 'zero' philosophy.
===== begin tldr =====
**We know that a human player that is othewsie WEAKER than LZ, (cannot …
-
release date: 2019-09
Expected:
- Jupyterlab-1.1.1,
- dashboarding:
- Anaconda Panel,
- Quantstack Voila, (in 64 bit only)
- not sure for Plotly Dash (but AJ Pryor is a fan),
- deep …
-
Hello
Thank you very much for your contribution to our research on real-time strategy games. How can I use the Bayesian method in the package? Is there a related introduction document?
From a g…