[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Hypothesis

Posted By: Marv Porten
Date: Thursday, 12 November 2009, at 12:50 a.m.

In Response To: Bot Comparison 100 / 500 games (Marv Porten)

Here's a possible explanation for the anomaly that increasing the ply (from 2 to 4 for GNUBG; from 3 to 5 for Extreme Gammon) doesn't necessarily lead to an improvement in the "Missed Double" category. Whereas in checker play looking ahead a couple of rolls greatly improves ones ability to select the best move, in many cases looking ahead a couple of rolls does not materially affect the decision to double. The reason for this is that (in many cases) the decision to double is based primarily on a determination of the long range outcome and only secondarily on what might happen in the next few rolls.

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.