I agree with what you say about UCT/MC and I think I made a similar post many months ago. Essentially I said, just as you, that UCT is closer to what humans do, it "works out" the particulars of the position.
I've always thought it odd that the approach advocated by many is based on static rules and intensive knowledge and the advocates of it call it the "intelligent" approach - or if you will they view it as more human like. As if humans don't really think - they just have it all in their head. I think what humans do best is reason. All animals have wonderful pattern recognition and memory about things, my bird talks and says things in context - he says "hello" when the phone rings and so on. But what seems to set our intelligence apart isn't that we have more of this, it's the fact that we can reason on it. UCT is far closer to this way of attacking a problem than looking up a set of patterns on the board and hand crafting hundreds of rules for telling you which move to play. And yet still it gets criticized as if it's a "dumb" approach (based mostly on the fact that it's not based on a database of knowledge which most classic programs essentially are.) In my view, the old classic knowledge based programs are like idiots that have a lot of knowledge of trivia, but no practical real world wisdom. I'm not criticizing modern programs like GnuGo and MFGO, they have kept up by adapting and adding lots of clever modules that in a sense "reason things out too", such as global search in MFGO - but UCT really takes things in a wonderful direction in my opinion. Of course there is nothing wrong with exploring all alternatives and methods. - Don Harald Korneliussen wrote: > In the thread "On average how many board updates/sec can top Go > programs do these days?" mingwu said of the way MC/UCT programs work > that he'd hardly call it intelligent. > I've thought (and argued elsewhere) that the MC/UCT approach is > fundamentally more "intelligent", in the sense of working more like > human intelligence apperas to do, than traditional game tree search. > How did humans learn to play Go? Well, they tried things, figured out > what worked and what didn't. And that is what a UCT seach does as > well... the main differences is that it's much worse at > generalization, and (for that reason) throws away most of what it's > learned from game to game. Still, it plays well with vastly less a > priori assumptions/knowledge than a traditional minimax-type search. > There is some, of course, but humans also rely on some a priori > knowledge according to the philosophers :-) > > To illustrate it a bit, I know of a game in which UCT's advantages are > even more striking. The game designer Nicholas Bentley recently won an > award for a game he calls Mind Ninja (silly name, I know, but bear > with me: it's a very interesting game). It's a generalized > connection/shape-building game, which begins with a negotiation phase, > where one player proposes the pattern to be built, and the other > decides whether he will be the "builder" or the "blocker". > There are extremely many possible patterns. On the regular board, > there are 127 points, which can be empty or coloured - the number of > colours are also part of the pattern rule. By a pattern rule, each > possible position must be either a win for the builder or not (the > blocker wins if there are no more legal moves and the builder hasn't > won). Each partitioning of positions into wins for the builder and > undecided positions/wins for the blocker can constitute a pattern > rule. > > Now, I believe MC/UCT can play this game, and play it well. (I'm > working on an implementation). It will adapt to pattern rules it has > never seen before, just like a human. It doesn't need any "hints" in > the form of evaluation functions, like chess programs use - in fact, > the nature of the game makes applying that approach pretty much > impossible. Like a human, it can consider a pattern rule, and try to > figure out which side has the advantage. Humans can certainly choose > rules which would give them an advantage (I imagine nim-like pattern > rules could work), but with trial and error, so could the program - > and the game has a way of balancing that advantage, too. > > Would a MN-playing program be intelligent? I'm pretty sure they would > have said so ten years ago, at least :-) But regardless, I think it > could illustrate the potential of this new algorithm. > _______________________________________________ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ > > _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
