Search results

You're kind of abusing the term minimax here. o_O There are two statements I'll take issue with: This is "technically" true, but misleading. You're making it sound as if the minimax player will always be able to predict what you're going to do. This isn't true; if you have multiple...

Went to the Madison, WI Gamestop pokemon tournament today (I'm saying where for a reason, wait a moment). Short version is that I won, and my g/f took 3rd, losing to me in the semi-finals. We got the shirt we wanted, and the person in 2nd place got the Pokemon Battle Revolution game they wanted...

Wow, it's really just amazing to have this level of discussion about pokemon AI here. ^_^ While I agree ideally, I don't really think that's their focus. As well it shouldn't be, they're trying to get Competitor out to the masses, not to the few of us who want to make AI for it. ;-) Having...

Serebii does most of it, but there was some list that had numbers for each effect, and this is kind of what I was looking for, so I know whether the effects of certain moves are slightly different. For example, the sleep from rest is different from the sleep from spore, the poison from toxic...

There was a list somewhere of all the moves in the game, with game-internal information about each move (such as move priority, effect number, type, power, accuracy, and a bunch of other unknown fields). I can't find it anymore. Does anyone know where this is/what I'm talking about?

This is more like Q-learning/reinforcement learning (and I think you can also do things like this with normal neural nets, maybe Brain will have more to say on this?) As far as modifying its team, I think there are a number of interesting ways to do it. The way I'm working on, every time the...

Well, I think that, realistically, you're going to have to have some function that knows the rules of pokemon anyway. I suppose it's possible to do it completely through learning, and observing others games, but I know for the method I'm using I'm going to have to write most of a pokemon rules...

There's a really big difference. First, Reversi (and checkers/chess/etc) are games of perfect information. Pokemon is not. As such, you have to deal with far more than the 4/9/14 moves that are possible each term. This doesn't even begin to get into chance nodes, etc., which is another...

tl;dr part 3 LOL at luring economists out of nowhere. ;-) I'm just a long-time lurker. I don't know much about actual competitive pokemon, but I know a little bit about this, and Smogon is amazing for battling information. One thing I noticed in a second-readover of the thread, probably not...

tl;dr part 2 A great start, McGraw. It turns out I was having an interesting discussion about this with my g/f today. ;-) Before I get too far into talking about this, an article I want to throw in that fits into the idea of "profiling your opponent" is from Sirlin (again)...