Re: [Computer-go] OT (maybe): Arimaa bot notably stronger

2015-04-23 Thread Stefan Kaitschick
The funny thing is, that in computer go there are no goals, except winning. And therefore, the reason for a win cannot be determined. One crude measure might be to use stronger attacking moves in the playouts, when the winrate is low. unrelated: Does anyone know if the successful Arimaa bot

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Ingo Althöfer
Hi Darren, wow, thanks for the interesting news. On http://arimaa.com/arimaa/challenge/2015/showGames.cgi the games can be replayed. bot_sharp won all three games of round 1 and also all three games of round 2. In round 3 the bot beat the medium strength human, but lost to the weak one

[Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Ingo Althöfer
Sorry for wrong format in my posting. Here it is in txt-style. ***  Hello,   I had just looked at the ratings of the players at the start of the match:   2623 Browni...    2262 Harvetsnow 2181 ChessandGo   My term weak was meant only relative to the

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Justin Blank
Sorry, I neglected to specify: Chessandgo = Jean Daligault. On Thu, Apr 23, 2015 at 1:54 PM, Justin Blank justin.bl...@gmail.com wrote: Not sure where you got that idea. Chessandgo won several Championships. He's probably the second or third best human player, after Matthew Brown, and

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Justin Blank
That's a reasonable assumption, but those are gameroom ratings, which Arimaa players typically don't weight too heavily. It's too easy to get inflated ratings playing against bots that you can reliably beat, and the results have no history, unlike WHR ratings, which means that playing repeatedly

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Justin Blank
Not sure where you got that idea. Chessandgo won several Championships. He's probably the second or third best human player, after Matthew Brown, and possibly Hirohumi Takahashi, who disappeared after the 2012 championship. 1. http://en.wikipedia.org/wiki/Arimaa#World_Championship 2.

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Stefan Kaitschick
On http://arimaa.com/arimaa/challenge/2015/showGames.cgi the games can be replayed. That's nice. Arimaa looks like a rather tedious game to actually play though. I have never played it, but the idea seems to be to take the opponents camel hostage at a sinkhole, and then go bossing around

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Ingo Althöfer
Hello, I had just looked at the ratings of the players at the start of the match: 2623 Browni... 2262 Harvetsnow 2181 ChessandGo My term waek was meant only relative to the two other humans in the competition. Of course, I would likely have no chance against all three. Regards,

Re: [Computer-go] Arimaa bot notably stronger

2015-04-23 Thread Janzert
For reference here's a much better set of Arimaa ratings using WHR http://home.scarlet.be/~woh/whr/whrh.htm Currently: Rank Player Rating 1browni3141 2822.3 4chessandgo 2499.8 13 harvestsnow 2323.3 Also of note is browni3141 just won the world championship for this year and

Re: [Computer-go] OT (maybe): Arimaa bot notably stronger

2015-04-23 Thread Petri Pitkanen
I played few games against bots in arimaa.com and they seemed to react. I think the eval can be used for that? I did not find the game interesting. Just being hard for computers does not make it fun. So I quit playing after few games Petri 2015-04-23 18:32 GMT+03:00 Stefan Kaitschick