-
### Description
A better way to train turn based, *searchable* environments I think is missing from RLlib. The AlphaZero implementation is there yes but it seems very restricted. In general, MCTS is …
-
```[tasklist]
### Development tasks
- [x] https://github.com/sopra-fs23-group-40/Server/issues/55
- [x] https://github.com/sopra-fs23-group-40/Server/issues/56
- [x] https://github.com/sopra-fs23-grou…
-
For this part of the assignment, you have to create mockups for the whole user interface of your future web application. Describe the flow among the different screens of your application. Specifically…
-
内容
---
ユーザ対CPUの機能を実装してください。
達成条件
---
- [ ] アルゴリズム
## Random Player
まず手元のピースの中からランダムに置くピースを選択し、そのピースが配置可能な位置を取得する。(確かAvailablePlayerMovesみたいな配列を返す関数があった気がする)
そして配置可能な座標が入っているその配列からランダ…
-
-
TLDR; Implement dynamic move temperature like KataGo has. Experiment with different parameterizations, validate through experimentation.
---
My first AlphaZero run for c4 was slow to recognize t…
-
Right now, the softmax temperature is part of the neural network cache key. This was to avoid the cost of either performing a softmax operation on cache-retrieved evaluations or storing a softmax'ed …
-
*Perkelta iš discord*
@Cluelessas
Už pačio res turėjimą padaryti savaitinį mokestį ir jai ten 3 savaites iš eilės nesumoka už res (nebūtinai 3) res būtų pašalinimas, taip būtų išvengiama nenaudo…
-
Under `https://docs.software-challenge.de/allgemein/spiel`
1. Hyperlink "...Das Spiel der Software-Challenge Germany 2023 ist [Hey, Danke für den Fisch](https://docs.software-challenge.de/spiele/p…
-
### known issues:
1. chat and map will stop syncing between both players after playing for a set amount of time
2. players will not be able to place their pieces when issue 1 happens
### scenario…