Search found 893 matches

by AlvaroBegue
Tue Feb 19, 2019 11:57 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: categorical cross entropy for value
Replies: 7
Views: 346

Re: categorical cross entropy for value

I understand your concern, but I suspect in practice it won't matter. Looking at enough examples, the network can figure out that strong positions for white end up being either won by white or draws the vast majority of the time, even if it doesn't know that loss < draw < win.
by AlvaroBegue
Mon Feb 18, 2019 5:51 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: categorical cross entropy for value
Replies: 7
Views: 346

Re: categorical cross entropy for value

You don't propose an alternative, so it's hard to discuss pros and cons. However, these matters always come down to an empirical question. Having a scheme that produces distinct probabilities for the outcomes has the advantage that they can be combined however necessary when we use the network. For ...
by AlvaroBegue
Thu Feb 14, 2019 10:55 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: A0 subtlety
Replies: 25
Views: 1293

Re: A0 subtlety

Was a set of input planes representing a position at time (t - T + 1) oriented for the side to move at time t or time (t - T + 1)? All the history input planes are oriented for the current side to move, i.e. at time t. Would it matter if they weren’t? (That’s a serious question btw) I suppose not b...
by AlvaroBegue
Mon Feb 11, 2019 12:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: delete informant symbols for one color in pgn
Replies: 10
Views: 463

Re: delete informant symbols for one color in pgn

Here's a hacked-together Perl implementation: cat kk.pgn | perl -ne ' if (/^\[/ || /^$/) {print; next;} for $token (split " ", $_) { if ($token =~ /^[1-9]/) {$wtm = 1; print "$token "; next;} if ($wtm == 0) {$token =~ s/[!\?]//g;} print "$token "; $wtm = 1-$wtm; } print "\n" ' That removes markings ...
by AlvaroBegue
Fri Feb 01, 2019 11:22 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: So, how many of you are working on neural networks for chess?
Replies: 22
Views: 1537

Re: So, how many of you are working on neural networks for chess?

smatovic wrote:
Fri Feb 01, 2019 6:03 pm
Myself still ponders on how to combine an AlphaBeta searcher with gpgpu ann eval...
I have a promising idea for that, but I don't really have the time to implement it. My job, my wife and my kids are getting in the way of the really important stuff! :)
by AlvaroBegue
Mon Jan 28, 2019 12:54 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI question
Replies: 11
Views: 798

Re: UCI question

If you have access to Perl, you can have much more control, like this:

Code: Select all

perl -e '$|=1; print "uci\nisready\nposition fen r1bk1n1r/pp1n1q1p/2p2p1R/3p4/3PpN2/2NB2Q1/PPP2PP1/2K1R3 w - -\ngo movetime 1000\n"; sleep (2); print "quit\n"' | stockfish
by AlvaroBegue
Thu Jan 24, 2019 2:09 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Lazy move generation and move ordering
Replies: 7
Views: 720

Re: Lazy move generation and move ordering

You should concentrate on move ordering, because it's much more important than lazy generation. In my engine RuyDos I roughly do the following: * generate captures and sort them by MVV/LVA, * filter the losing captures (SEE<0) and save them for later, * generate non-captures and sort them by history...
by AlvaroBegue
Tue Dec 18, 2018 5:51 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Policy training in Alpha Zero, LC0 ..
Replies: 26
Views: 1203

Re: Policy training in Alpha Zero, LC0 ..

Input of a training example consists of position value plus probability for each legal move And how do you know the probability for each legal move from a PGN? You don't. In the normal training of AlphaZero and LC0, your training samples are written at the end of a search, and the visit count of ea...
by AlvaroBegue
Tue Dec 18, 2018 5:45 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Policy training in Alpha Zero, LC0 ..
Replies: 26
Views: 1203

Re: Policy training in Alpha Zero, LC0 ..

Henk, you are adding noise to the conversation.

What you are describing is the sort of reinforcement learning that was described in one stage of the initial AlphaGo paper. We are not talking about that here.
by AlvaroBegue
Tue Dec 18, 2018 4:35 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Policy training in Alpha Zero, LC0 ..
Replies: 26
Views: 1203

Re: Policy training in Alpha Zero, LC0 ..

Input of a training example consists of position value plus probability for each legal move And how do you know the probability for each legal move from a PGN? You don't. In the normal training of AlphaZero and LC0, your training samples are written at the end of a search, and the visit count of ea...