It all comes down to having a reasonable way to search the MCTS’s tree. An elegant tool would be wonderful, but even something basic would allow a determined person to find interesting things. When I was debugging SlugGo, which had a tree as wide as 24 and as deep as 10, with nothing other than an SGF reader displaying the board and a window showing me numerical values encoded as comments, it was fairly easy to understand why the program made its choices. Translating from those numerical values and sequences into “language based evaluations” of the type suggested below became easy for me and did aid in making the program stronger. Initially SlugGo beat me consistently, but after a month of looking at those sgf files I internalized enough of how SlugGo evaluated things that I never lost a game to it again, even after modifications to the evaluation function.
In all honesty, I’ve had Go teachers who’s comments in lessons (not your choice; ‘this’ is the best move) were less useful to me than being able to explore so many different evaluated branches of the tree. Cheers, David G Doshay ddos...@mac.com > On 31, Mar 2016, at 2:09 PM, uurtamo . <uurt...@gmail.com> wrote: > > Major changes in the evaluation probability could likely have a horizon of a > few moves behind that might be interesting to more closely evaluate. With a > small window like that, a deeper/more exhaustive search might work. > > s. > > On Mar 31, 2016 10:21 AM, "Petr Baudis" <pa...@ucw.cz <mailto:pa...@ucw.cz>> > wrote: > On Thu, Mar 31, 2016 at 08:51:30AM -0500, Jim O'Flaherty wrote: > > What I was addressing was more around what Robert Jasiek is describing in > > his joseki books and other materials he's produced. And it is exactly why I > > think the "explanation of the suggested moves" requires a much deeper > > baking into the participating ANN's (bottom up approach). And given what I > > have read thus far (including your above information), I am still seeing > > the risk extraordinarily high and the payoff exceedingly low, outside an > > academic context. > > I think we may just have a different outcome in mind. To illustrate > where I think my approach could work, that could be for example > (slightly edited): > > > White Q5 was played to compel Black to extend at the bottom. > > If Black doesn’t respond, White’s pincer at K4 will be powerful. > > in > https://gogameguru.com/lee-sedol-defeats-alphago-masterful-comeback-game-4/ > <https://gogameguru.com/lee-sedol-defeats-alphago-masterful-comeback-game-4/> > > > Sure, it seems a bit outrageous, and for initial attempts, generating > utterances like > > > White 126 was a very big move which helped to ensure White’s advantage. > > is perhaps more realistic (though many of these sentences are a bit > of truisms and not terribly informative). But I'm quite convinced that > even the first example is completely plausible. > > (But I'm *not* talking about generating pages of diagrams that > describe an opening position in detail. That's to ponder when we > get the simpler things right.) > > -- > Petr Baudis > If you have good ideas, good data and fast computers, > you can do almost anything. -- Geoffrey Hinton > _______________________________________________ > Computer-go mailing list > Computer-go@computer-go.org <mailto:Computer-go@computer-go.org> > http://computer-go.org/mailman/listinfo/computer-go > <http://computer-go.org/mailman/listinfo/computer-go>_______________________________________________ > Computer-go mailing list > Computer-go@computer-go.org > http://computer-go.org/mailman/listinfo/computer-go
_______________________________________________ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
