Arimaa

hyperpape · #1

Who here plays? I'd heard of it before, but yesterday was the first time I played, and right now I feel addicted.

I looked at some high level games, and go to reexperience the feeling of a 30k looking at a professional game.

daniel_the_smith · #2

hyperpape wrote:

Who here plays? I'd heard of it before, but yesterday was the first time I played, and right now I feel addicted.

I looked at some high level games, and go to reexperience the feeling of a 30k looking at a professional game.

I've played before. I think I got up to around 1500 or so on the rating system playing bots, but I don't know if I ever got as far as playing humans.

I even have a half-written arimaa bot sitting on my computer. Sigh, things I will never have time to complete...

Mivo · #3

I played it for a while, a few years ago, but didn't get very far with it. I also didn't know anyone else who played and it was before they had a physical set available, so my interest in it kind of fizzled. Plus, my usual problem: I generally find it hard to get interested in any other strategy games ever since I learned (about) Go.

hyperpape · #4

I just crossed 1400, the 1500 bots are still way too tough for me. And I've yet to play a human, which is just a wee bit ironic.

Laman · #5

another beginner here, played 20 games in the last 2 weeks. it is nice to learn something completely new again, to suck so much that you improve just by playing and random messing around

but lack of human opponents is troublesome, i also played only one human so far

Numsgil · #6

Looks interesting.

Are there any monte carlo bots? I'd imagine that would be the way to make the strongest computer player.

daniel_the_smith · #7

Monte Carlo does pretty bad at Arimaa, actually! Random play moves the rabbits way too much. It's basically only our own game that MCTS has done really well at. :sad:

[EDIT: Last I heard, anyway. I don't know if anyone has tried the statistical learning techniques (other than RAVE, which IIRC did not work well for Arimaa) that work for go.]

Mike Novack · #8

Numsgil wrote:

Looks interesting.

Are there any monte carlo bots? I'd imagine that would be the way to make the strongest computer player.

Not completely understanding the situation. The game "Arimaa" was devised with all currently known methods for computers to play games in mind with the intent that these be ineffective for Animaa.

Notice that before it was thought of and applied far from clear that MCTS would be the best currently known way for computers to play go. That was after many years of trying other things that weren't able to play a strong game. Well there may or may not be a way for computers to play a strong game of arimaa but hasn't been discovered yet.

daniel_the_smith · #9

Yeah, Arimaa was designed before MCTS hit the big-time with go. MCTS just happens to not work well for Arimaa (just as it doesn't work well for chess, for much the same reasons).

hyperpape · **#10**

It's not obvious to me that bot progress in Arimaa is worse than bot progress in Go. Of course it's complicated because both humans and bots are very new to the game, so both are making progress at the same time, and you don't have a fixed target to aim for.

And you can't imagine bots placing so highly on a list like this for Go (even given the disparity in number of players, imho) http://arimaa.com/arimaa/gameroom/topRatedPlayers.cgi

daniel_the_smith · **#11**

I agree with that, too. Traditional chess engine techniques have worked fairly well in Arimaa, but not quite well enough to stay ahead of the humans. Arimaa is young and has few players, so it's actually a little surprising how well the humans have done!

Numsgil · **#12**

Interesting. I'd have guessed Monte Carlo was the way to go.

Maybe Monte Carlo works well with go because the final board position isn't a binary win/fail, but a gradation of points for either side. Chess and Arimaa are both strictly win/lose (draw). I'm not sure why that'd matter, but that's the only thing I can think of that is really fundamentally different between the games. I'm sure there's some really interesting fundamental truth about Monte Carlo creeping in here somewhere, but I can't figure it out.

daniel_the_smith · **#13**

One reason it works well in go (IIRC) is because random errors roughly cancel out. MCTS doesn't care about the margin of victory at the end, it aggregates binary win/lose events into a probability of winning.

Numsgil · **#14**

daniel_the_smith wrote:

One reason it works well in go (IIRC) is because random errors roughly cancel out.

That's more or less what I meant. Like, a random walk from a given even position proceeds to a slightly less even position, from which a random walk might walk back to the even position. At the end of the game there's a clear outcome but before then the game has this inherent gradation. Or to put it in terms of selection, the fitness landscape is smooth, with few steep gradients. Versus Chess and Arimaa where things could be going fine until they're not and you lose. So the fitness landscape has all these horrible cliffs and canyons everywhere.

I don't know if I actually believe that or not. Just an interesting thought.

Mivo · **#15**

I actually wonder why the game isn't more popular. Thanks to this thread, I picked up the iOS Arimaa app and had a good time playing against the bot (last time I looked into Arimaa the only bot I found was David Fotland's). Is the pulling/pushing/freezing too complicated for people, or is it the stale presentation of the site? Or is it just hard for new board games without commercial backing or a very long history to take off?

It seems off that a game that is designed to be hard for computers has numerous bots in its list of the world's top 100 players.

robinz · **#16**

I'm quite a keen Arimaa player, although not all that good - rating currently a shade below 1800. (In fact, while it is heresy to state this here, I actually much prefer it to Go. I've barely played Go in the past 6 months, while I've been playing quite a lot of Arimaa online.) I also struggled on the bot ladder to start with, but I later discovered the "autopostal" feature which automatically pairs you with human players of similar rank, for turn-based games. I've actually won a largeish majority of these games, which shows either I'm better than I thought or that the rating system doesn't work that well (but of course it's getting harder now as my rating goes up and I get paired against stronger players).

My personal view is that Arimaa is the only abstract game invented in the past 50-100 years which has a chance - indeed, in my opinion, deserves - to be played and studied for as long as chess and go have. I don't expect it to happen though - primarily because, as Mivo says - it's hard for abstract games to really get a foothold in the market without a major publisher backing it. And also because most of those who would be prepared to dedicate most of their time to a game like this already do so with chess or go, and don't want to devote the same amount of time to a different game :lol:

daal · **#17**

Mivo wrote:

I actually wonder why the game isn't more popular.

Mivo wrote:

I generally find it hard to get interested in any other strategy games ever since I learned (about) Go.

This sums it up for me. Go is a classy game with a tradition thousands of years old, (and chess apparently has a long tradition as well ;-)

) - but a game like this in comparison seems too fly-by-night for it to be worth investing the time necessary to get good. Look what happened to that online game that was similar to go but with invisible fields or something (badoo?) Gone with the wind.

Mivo · **#18**

robinz wrote:

And also because most of those who would be prepared to dedicate most of their time to a game like this already do so with chess or go, and don't want to devote the same amount of time to a different game :lol:

Speaking of which ... just last night I was discussing with my better half(*) whether it's necessarily a good approach for me to spend a lot of time and effort on one game in the (vain or otherwise) hope to become proficient at it, instead of putting time and effort into a variety of games, aiming to become merely "decent" at them, but in return experience a wider spectrum of intellectual challenges and input.

For me personally, I think the best time I've had with Go was when progress came easy, when there was a lot to explore and discover, learning every time I sat down with the game, when new knowledge came naturally. In the SDKs, that changed, and making further progress seems, at least for myself, to take substantially more effort for substantially less gain. Or differently put, I think 80%+ of what I learned about the game, I learned in the first half year. The years that followed yielded the rest, and the effort/progress ratio only continues to get worse. I think I even enjoy "learning" more than I do "playing".

At the same time, if I'm honest with myself, I'm not even having more fun. There are still the times when I learn something new or feel that I get stronger, but it's not as frequent anymore. And actually, for the most part I'm "OK" with being able to play a game of Go, doing tsumego, and watching other people play without being confused about what's going on all the time, and I enjoy the Go community and all the cultural aspects. But do I need to get to the dan levels? No doubt, I would like to, but on the other hand it seems to mostly require hard work for relatively slow progress. Above all, when I catch myself thinking along these lines, it strikes me that I seem to be falling into the "destination more important than the journey" trap. I don't think I would enjoy the game more if I was 1dan vs. somewhere in the mid SDKs, but I am almost certain that I probably won't enjoy putting in the required effort for increasingly less noticeable gain.

I guess what I'm getting at is the question (asked to myself) whether it wouldn't make for a better intellectual experience to pick up a new game every year and learn it, enjoying the thrill of progress and fresh insights, and seeing how far one can get while the balance between progress and effort still favours progress, when learning is exciting rather than a grind.

An invention-wise young game like Arimaa, to use it as an example, also has the benefit that there aren't hundreds of years of research and all these people who were already stronger at age 10 than you know you can ever hope to even remotely become. There's the attraction of potentially new discoveries, the feeling of possibly being part of the unfolding of a game, the fumbling around with others instead of the looking up to the masters and feeling hopelessly and eternally inferior. More than forever only following footsteps.

Of course, that's also a downside. A game like Go offers an endless pool of source material, high level matches to learn from, and a wealth of philosophical and cultural beauty (then again, that is, at the end of the day, just "fluff": there is nothing inherently spiritual about a board game; it's just being projected on the game, but at the core, Go is only a highly abstract strategy game), and the knowledge that the game is one that has lasted and will last throughout time, and that won't just fizzle out because it doesn't have enough longevity.

Anyway, just some poorly structured thoughts.

(*) Fine, so I was philosophizing and she listened, nodding here and there!

palapiku · **#19**

Go would probably not become popular if it were invented today.

CSamurai · **#20**

Coincidentally, I'm working on a puzzle game for the MIT mystery hunt.. and one of the puzzles is Arimaa based. What are the odds?

Any good players out there able to solve particular 'x to play and win by' problems?

Arimaa

Who is online