Hi,

I only had a short cross-read, so my impressions are very subjective.

>   I don't think the paper is written very clearly, 

Right. The authors are either not willing are not able
to formulate their findings in simple words.


>   * Do you think the results feel plausible?

Yes.
 
>   * Do you think it contributes anything actually immediately useful?

I think they give "us" old wine in new bottles.

The following sentence from short Section 3 (on p.5 middle) feeds my doubts:
> Quite surprisingly, very powerful programs have already been developed 
> and tested in practice against human players (see [11]), however the 
> action-evaluation algorithms used in these software are purely heuristic
> and no theoretical foundation is presented to explain their success.

Ref [11] is the wellknown paper by Kocsis, L. and Szepesvari, C. (2006).

>From Remi Coulom they only mention the 2007-workshop report on use of Elo
ratings...

I was not impressed - or their really good findings are hidden somewhere
(either in the 54 pages or in the forthcoming papers announced).

Ingo.


-- 
Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to