Author: Robert Hyatt
Date: 06:16:03 03/20/98
Go up one level in this thread
On March 20, 1998 at 03:16:55, Bruce Moreland wrote: > Ed Shroeder wrote: >>In fact the only thing I would have done (please read my previous >>posting) is that I have taken advantage of the fact that I know NOW >>*HOW* SSDF testers do their testing. >> >>I don't want to be a part of such a development but this kind of >>things is happening right before our eyes since a few years. > >This is a different thing entirely. Right now the SSDF guys might do >one game or one hundred games, or do a game then exit your program and >not exit the other guy's program. I don't know how they do it, but if >they are doing it weird perhaps they should think about standardizing. > >Maybe it should be the programmer's responsibility to make sure that >learning data gets saved at some point during or after a game, and it >should be the SSDF's responsibility to try to do games in match units of >at least perhaps ten games, on the same machines, between the same >programs, although it would be fine to interrupt the match at any point, >they wouldn't have to do the games all at once. > >You don't need to resign anything. This stuff is good for everyone, >it's just a matter of making sure that programs are robust enough to >handle real world situations, and the SSDF people are consistent enough >that the results are approximately what they should really be. > >I'm sure other stuff like this will come up. We're not going to be >using the same approaches forever, we are all going to do more stuff >better, and this learner stuff is just an example of this -- the >competitive domain just got more complex, which means more interesting >and more fun for everyone, especially us. > >bruce This is a flaw in using SSDF and auto232 for the basic testbed. As Bruce and I both know, ICC is a much more hostile environment for a chess program to "live" in. My learner works *regardless* of what anyone does to it. IE terminate a game in the middle, it still learns. Terminate and restart the game 10 times? Doesn't affect me in the least now, I still learn just I should, when I should, and what I should. Ie it boils down to my being responsible for what I do and how I do it. I'm not going to depend on my opponent to behave in a specific way so I can learn properly.. I'm going to assume my opponent will try everything possible to break my learning algorithm, and then I'll fix every hole I find. Ergo, fix things, don't wait for the SSDF to fix them. That way you can not be surprised, and there's no excuses to be made... I've always thought it was stupid for auto232 to play a series of white games. That is *not* the way real chess is played. So their flipping colors when using Fritz, because the fritz autoplayer does this, seems perfectly natural to me, since that is the way it ought to be. But for crafty, it doesn't care. But for every game you beat it with white, don't be surprised if it tries the same opening when *it* plays white...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.