twitchess

A toy implementation of neural network chess written while livestreaming.

Stream

https://www.twitch.tv/tomcr00s3

Usage

$ pip install -r requirements.txt
$ python play.py   # runs web server on http://localhost:5000

TODOs

Roll out search beyond 1-ply
Make trainer multi GPU
Train on more data
Add RL self play learning support

Implementation

twitchess is a simple 1 look ahead neural network value function. The trained net is nets/value.pth. It takes in a serialized board and outputs a range from -1 to 1. -1 means black wins, 1 means white wins.

Serialization

We serialize the board into an 8x8x5 bitvector. See state.py for how.

Training Set

The value network is trained on 10M board positions from http://www.kingbase-chess.net/

Zero Knowledge Chess Engine (liar!)

Memo

Establish the search tree
Use a neural network to prune the search tree

Definition: Value network V = f(state)

What is V?

V = -1 black wins board state
V = 0 draw board state
V = 1 white wins board state

Should we fix the value of the initial board state?

What's the value of "about to lose"?

Simpler:

All positions where white wins = 1
All positions where draw = 0
All positions where black wins = -1

State(Board):

Pieces(2+7*2 = 16):

Universal
- Blank
- Blank (En passant)
Pieces
- Pawn
- Bishop
- Knight
- Rook
- Rook (can castle)
- Queen
- King

Extra move:

To move

8x8x4 + 1 = 257 bits (vectors of 0 or 1)

Download training data

$ wget http://kingbase-chess.net/download/599 data/kb2018.zip

vyasn30/twitchess