Search found 28 matches

by ddyer
Sun Jan 28, 2018 1:34 am
Forum: Machine Learning
Topic: there must be a better way
Replies: 1
Views: 3316

there must be a better way

Regarding using neural nets for games. Doesn't it seem odd that we take a domain of perfectly discreet objects - black stones, white stones, squares on the board, and re-represent it as floating point numbers with thousands of parameters? It seems like there ought to be an approach that uses similar...
by ddyer
Wed Jan 24, 2018 9:09 pm
Forum: Machine Learning
Topic: some newbie notes about Neural Net AIs
Replies: 1
Views: 593

Re: some newbie notes about Neural Net AIs

I've been using Hex on a 7x7 board as a test bed. My basic approach is to use the visit counts at top level of a MCTS search as the target; make the network learn what the visit count should be, and use that prediction to weight the random playout phase of new searches. I understand this is the basi...
by ddyer
Thu Jan 11, 2018 6:07 am
Forum: MCTS
Topic: MCTS search visialization
Replies: 1
Views: 3269

MCTS search visialization

Here's a visualization of how a MCTS search evolves.
https://www.youtube.com/watch?v=FvRSxNLTg7U
by ddyer
Mon Jan 08, 2018 1:13 am
Forum: Machine Learning
Topic: About Leela Zero
Replies: 0
Views: 1073

About Leela Zero

Leela Zero is a green-field effort to reproduce and explore the methods used by Alphago zero which I've been following. http://zero.sjeng.org/ It seems to be having some success, and it's open source. The general archecture it uses is a NN similar to the one described for Alphago zero, which guides ...
by ddyer
Sun Jan 07, 2018 10:29 pm
Forum: Machine Learning
Topic: some newbie notes about Neural Net AIs
Replies: 1
Views: 593

some newbie notes about Neural Net AIs

Inspired by Alphago and Alphazero, I've started exploring NN methods for game AIs, and neural nets in general. For starters, I found that there are no good forums for general discussion, hence this Machine Learning section, and this post. I'm kind of a bottom-up guy, I like to understand the stuff I...
by ddyer
Wed Oct 18, 2017 12:23 am
Forum: MCTS
Topic: MCTS AI development for "Tintas"
Replies: 0
Views: 1069

MCTS AI development for "Tintas"

Tintas is an abstract strategy game involving collecting colored tokens. https://boardspace.net/tintas/english/tintas-rules.html If you were trying to design a game to be a perfect vehicle for MCTS, this would be it. There are a maximum of 49 moves in the game, and usually the branching factor is 6 ...
by ddyer
Fri Jun 30, 2017 7:19 pm
Forum: General Discussion
Topic: Welcome to Download the computer-go-dataset (1,645,958 SGFs)
Replies: 3
Views: 1745

Re: Welcome to Download the computer-go-dataset (1,645,958 SGFs)

Thanks! FWIW, back in the stone age of computing I struggled to collect about 5K professional games.

One caveat about this data set; a lot of the file names include unicode characters, and a lot of the files contain UTF-8 text.
A lot of tools will behave badly because of this.
by ddyer
Fri Jun 30, 2017 7:15 pm
Forum: General Discussion
Topic: Insight: the best evaluator depends on the search depth
Replies: 2
Views: 1272

Re: Insight: the best evaluator depends on the search depth

The depth is fixed, but maybe you missed the point. For the same pair of evaluators, the one tuned to win at 1-ply does much worse if the search is deeper. For the actual evaluators I was testing, the actual playing depth is 5 or 6 ply, and it just didn't work to use a 1-ply search as a proxy to sel...
by ddyer
Tue May 30, 2017 7:56 pm
Forum: General Discussion
Topic: Insight: the best evaluator depends on the search depth
Replies: 2
Views: 1272

Insight: the best evaluator depends on the search depth

I've been experimenting with a genetic algorithm to improve the evaluator for my Hive implementation, and because I wanted fast results, my simulated games were run with a 1-ply search; Ie; using the evaluation directly, without any deeper search. After various experimentation, I developed a procedu...
by ddyer
Thu Jan 14, 2016 8:22 pm
Forum: General Discussion
Topic: Paper describing "Sharp" the program that won the Arimaa Challenge
Replies: 1
Views: 2542

Paper describing "Sharp" the program that won the Arimaa Challenge

http://icosahedral.net/downloads/djwu20 ... _color.pdf

Among other things, it has an excellent description of the develop process over time.