Re: computer-go: Temporal Difference Learning

[Peter Drake <drake@xxxxxxxxxx>]

On Monday, July 14, 2003, at 03:21  PM, Markus Enzenberger wrote:

> > Are there any major deviations from this plan within TD
> > programs? Specifically, is anyone:
> >
> > ...doing more than 1 ply lookahead?
> there is an algorithm called TDLeaf, but I am not convinced
> that it is useful.
A quick web search found a paper by Baxter, Tridgell, and Weaver.  Is 
this the canonical one?

Also, can you say why you're not convinced this is useful?

> > ...using TD to learn to answer more specific questions,
> > such as, "can these two chains connect"?
> NeuroGo in its most recent version uses local connectivity
> and single-point eyes as additional outputs that are
> trained with TD. I will present a paper about this at
> ACG2003 which takes place together with the Computer
> Olympiad in Graz/Austria in November.
So when and how do those of us stuck stateside get ahold of it?  :-)

Thanks,

Peter Drake
Assistant Professor of Computer Science
Lewis & Clark College
http://www.lclark.edu/~drake/