BGonline.org Forums

Training bots on alternative starting positions

Posted By: Timothy Chow
Date: Friday, 27 November 2009, at 3:07 a.m.

In Response To: Training bots on alternative starting positions (Frank Berger)

But during the training the difference got smaller and it seemed to converge :(

Is that a frownie face? I don't see why this should be a disappointing result, even if it survives a stiffer test of significance. What it suggests is that Nackgammon is not all that different from backgammon.

So I would be interested in seeing what happens if you train a bot on Snake. If such a bot behaves just like an "ordinary" bot, both on ordinary positions and on deep-backgame positions, then I would indeed be a little disappointed, because it would suggest that backgame strategy just can't be learned very well with current neural-net algorithms. However, I still think it's worth a shot, since backgames still seem to be a glaring weakness of current bots.

Messages In This Thread

• Training bots on alternative starting positions
  Timothy Chow -- Thursday, 26 November 2009, at 4:56 p.m.
  • Training bots on alternative starting positions
    Frank Berger -- Thursday, 26 November 2009, at 8:23 p.m.
    • Training bots on alternative starting positions
      Timothy Chow -- Friday, 27 November 2009, at 3:07 a.m.
      • Training bots on alternative starting positions
        Frank Berger -- Friday, 27 November 2009, at 7:58 p.m.
      • Training bots on alternative starting positions
        playBunny -- Sunday, 29 November 2009, at 11:24 a.m.