-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The results look more reasonable now, 11/11 usually indicates a clear superiority.
Still a 70 ELO superiority of housebot but that is within a believable range, it could turn around if we give then enough time. >> How many all-as-first moves are you considering? > > 7/8 of the moves until the end of the simulated game. This will vary > from random sim to random sim if the game lengths are different. It > also uses passes as part of the game length calculation. A result of 0 > moves is replaced with 1 move. Out of that list, only moves for the > color to play are considered, and only if the opponent didn't play in > that spot first. > > >> What are the possible >> differences? >> >> 1. eye rule > > I thought we beat this one to death. It's an eye if all 4 neighbors > match the color to play (or are off the board), and one of the following > are true: > 1. It's a corner and the diagonal position is not an enemy stone > 2. It's an edge and neither of the diagonal positions are an enemy stone > 3. It's in the center and no more than one of the diagonal positions are > an enemy stone > >> 2. number of all-as-first moves (7/8 for me + 1) > > Explained above, but I realize I got one small detail wrong. It's > max(1, min(1000,7/8*moves)) I think mine comes down to that in 99.99% of the cases. >> 3. quality of RNG > > Mersenne Twister (generic implementation available free online) Same as AnchorMan >> 4. correctness of random move selection strategy. > > Pick a random empty position. If illegal or eye-filling, remove from > consideration the list and repeat. Same basic idea. I start by taking all filled points and removing them from consider but I reinitialize after a capture. >> 5. depth at which you stop a game. (about 1000 moves for me.) > > I never stop a game. They always play out to the end. I put an upper > bound of 1000 on the moves to keep for AMAF, but the random game will > continue until it's completed. > >> 6. stopping rule. (both program have no non-eye filling moves.) > > I use that same rule. If a program hits that condition, it's forced to > pass. Play then continues until there's two passes in a row. Each pass > is considered a move for purposes of AMAF counting. Exactly the same for me. >> I just thought of something. I think I initialize the statistics array >> with 1 draw per move as a cheesy way to avoid divide by zero error. >> Could this be affecting the performance? Perhaps at low levels like >> this it has a noticeable effect? Would it make the program especially >> vulnerable to an identical program that doesn't do this? > > Draw? Is that a valid outcome of your random simulations? I would > assume your counters would be integers (and a draw value of 1/2 wouldn't > work). A win is 1 and a loss is -1 so I set all the game counters to 1 - which is like a draw. Of course in the simulations a draw can never happen. > > Curiously enough, you found a potential bug in my amaf implementation. > Divide by zero could occur (but doesn't seem to) > > Also, we may differ on our handling of ko. I use simple ko rules in my > playouts. I think that differs from other implementations that don't > use ko rules at all. I don't use any super ko logic anywhere in my bot > at the moment. I use only simply ko in the simulations. But outside the simulation I honor positional superko - so AnchorMan will never make an illegal move due to this. It's rare but some even some good bots don't do this and will lose a game once in a great while. It's probably worth less than 15 ELO to do this right. But that does mean GenAnchor_1k should be slightly stronger even though it doesn't appear that way now. - - Don > _______________________________________________ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFG/9HBDsOllbwnSikRAiQcAJ4h0kG+9iRvZSxwTBRnArYvRxrPQACg1PyM LrePa5fuhOXM/ChZbc/VYgA= =quAK -----END PGP SIGNATURE----- _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
