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

BGonline.org Forums

Why do bots have an even/odd effect?

Posted By: MaX
Date: Saturday, 15 August 2009, at 8:53 p.m.

In Response To: Why do bots have an even/odd effect? (playBunny)

playBunny wrote:Consider a position with strong contrast between the two sides, say white just finishing bearing in against a black anchor on the ace point, black with a full home.

A 0-ply evaluation will examine the completed bear-in, greatly in favour of white. A 1-ply will look at black's 21 rolls by making each move (a table crunch or a checker escape), inverting the resulting boards and evaluating each as if it were white in the unfavourable position.

I would say that after a chequer escape or a table crunch the favorite is the same player as before the move (i.e. white, the player with the bearin completed). I doubt a lot 1ply can get this wrong (as soon as 0ply gets it right). After you have inverted the board, white is the underdog but current white is your initial black. Or I'm missing something ...

Tom wrote: But if a bot is trained only to give accurate relative equities*, there is no guarantee that the absolute equities are at all close to being accurate. The even/odd effect is the difference in absolute equity viewing the board from one side as compared to viewing it from the other side.

I agree for the abs/rel equities issue (affecting cube decisions and maybe truncated rollouts), but still I don't see how this could generate an odd/even effect. Strong suspect now, am I missing something ?

Can you come up with a (possibly fake) example Tom ?

MaX.

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.