Why programmers should play Go | Hacker News Reader

Why programmers should play Go

railspikes.com

48 pointsjon_dahl51 comments

There's an interesting trend in the computer go community away from rule-based AI, and towards a more stochastic approach. This means that where the leading developers were all 1 dan or better (very talented amateurs, by the metric of the article), it's become possible for talented programmers with a relatively limited understanding of the game to make major contributions.

Go is a complicated game, and the human response to that is to create rules of thumb and build a strong intuitive understandings of different positions. It's arguably less elegant to analyze a position by running hundreds of thousands of playouts, but it's finally a problem that resembles real computer science (though most academic papers are still of the "we tried this, and it did okay" form). It's a pretty exciting time to be involved, and I'd suggest that anyone with even a passing curiosity check it out.

The archives of this mailing list has some good discussions about the stochastic approach:

http://computer-go.org/pipermail/computer-go/

Apparently the strongest computer AI is now able to beat 1k players consistently.

It's worth noting that it can beat 1k players on a 9x9 board. On a 19x19 board (the standard board size), they are both weaker and less consistent.

One of the strongest current programs, MoGo, recently beat a 5 dan player on 9x9 using a monte carlo approach.

http://en.wikipedia.org/wiki/Computer_go#Monte-Carlo_methods

I'd seen that, but I haven't seen any published results of 1k players being routinely beaten at 19x19. If I've missed something like that, I'd be grateful for a link.

agreed: 9x9 is to real go what checkers is to chess.

Yes chess isn't really interesting out of an AI-perspective since a supercomputer can beat a human with brute force + heuristics. In Go you'd need something more like real AI.

AI is search, brute-force or otherwise. There are a number of fascinating things that have come out of chess research on the way to grandmaster-level play, and even if the strongest chess players are all computers, that doesn't mean that it's suddenly become a worthless pursuit. The day that a perfect game of chess is played is the day that it gets boring, not before.

With respect to go, the current approach is completely brute force; it plays thousands of games at random, and uses the percentage of wins by one side to evaluate the strength of the position. The challenge is how to use the information you've already encountered in your searches to reduce the search-space to something manageable.

Chess could be interesting from an AI perspective if you limited the amount of brute force searching allowed. Computer chess and human chess are, by and large, completely different games.

I'm not sure I agree with this article about patterns in chess. The real masters think strategically as well as tactically, and use the structure of the pawns and pieces, combined with intuition, to guide their moves.

Ultimately, I suspect that go and chess fire up similar regions of the human brain.

They did MRI on Z.Polgar while she was playing. What they found out was chess thinking happens in a small region of the brain. The region is the same humans use when trying to recognize faces. Training modifies that machinery to recognize chess board positions too.

Huh, that surprises me. I would have expected a chess master to use more of her brain than I do (not that I know how my own brain would look all wired up, but I have trouble believing that I don't think much more narrowly about the game - kind of how people who actually understand math approach factoring a set of equations differently from people who need a set of rules they can follow closely).

Perhaps they only looked at what parts of their brain experienced chess players used more?

I generally advise people to play games that involve some randomness, like poker, gin, or backgammon, before moving on to games without. The reason being that almost all of life involves decision making in situations with some random factor, and humans are ill-equipped by biology for this. Most decisions that don't involve some unknown are relatively easy.

Go might be better for programming specifically (since randomness isn't a factor) but programming is a subset of life.

> Most decisions that don't involve some unknown are relatively easy.

Apparently you don't play go.

I play games for entertainment, not to mimic reality. Games with chance as a significant component aren't necessarily more entertaining than games requiring significant skill.

I would say the general population disagrees with you, which is why are games of chance a multi-billion dollar industry and only a few people can make a good living playing chess.

Well, other skill based "games" include the majority (or all?) of professional sports. If there were chance introduced at key points in the game (like drawing cards or flipping coins) I don't think anyone would watch baseball, football, or soccer.

Also, board games involving skill (e.g. chess and go) have stood the test of time over many chance games.

From http://en.wikipedia.org/wiki/Penalty_shootout_(football)#History

"Where a replay or playoff was not possible, ties were previously decided by drawing of lots. Examples include Italy's win over the USSR in the semi-final of the 1968 European Championship."

Chance is a huge element in sports. Otherwise nobody would watch them.

You're separating luck and skill as if games can't have both, and in the process entirely missing my point.

Clearly there is a spectrum, I'm not missing your point. In this spectrum though, chess/go/sport are at one end and gambling is at the other. Poker is in the middle.

Sport is probably closer to the middle than you think, especially where a team is involved. If you don't think an athlete can get unlucky, ask anyone on the Falcons. One teammate does something stupid and your team is out of contention for years.

Teams are generally businesses too, and just like most other businesses have a huge random element.

There is plenty of unknown in Go.

I must be playing it wrong. I've only played a couple times, but I thought it was a game of complete information. Perhaps I was taught incorrect rules.

Event thought it's a game of complete information I find it hard to foresee the moves of better players that kill my stones and capture more territory. And I can't console myself that he got the right card or had a good die roll. There is clearly a place for games of chance but the level of complexity in the go game tree makes it very interesting.

Well, not knowing your opponents' moves is common to every game. Otherwise its not a game.

Sure, but when I was five I found tic tac toe a challenge because I couldn't visualize the game tree. It's also a game of "perfect information."

Doesn't that agree with what I was saying? Games of perfect information are less skill intensive than some games that don't have it, because it's just a matter of visualizing the game tree, even though with chess or go the tree is too big.

If you are a fan of games, I heartily encourage you to give Go a good chance. It is both beautiful and profound, yet of broad practical benefit. Poker, though very different, is the only other game I know which is comparably elegant, applicable and revelatory.

I'm sure I'll spend some time on it some day.

I'd look at it more like chaos. Theoretically, there is absolutely nothing random in the shuffling of a deck of cards or the roll of dice, bracketing the human elements. It's just really really complex, so we call this chaos/random.

I mean unknown in the sense of unpredictability, chaos, combinatorial complexity.

I've never heard anybody complain that Go was too easy.

I meant more decisions in life in general. Go is certainly not easy.

If anyone wants to play me, I'm Alex3917 on KGS. I'm still a beginner though, only 12kyu.

http://www.gokgs.com/

Also, if you want a good introduction to the game, check out this K5 article:

http://www.kuro5hin.org/story/2002/7/11/62356/9269

That was the article that got me into Go in 2002. I list in and some other recommended Go resources here: http://www.davidsj.com/post.php?id=A72_0_1_0_C

thanks! Great collection of links.

Very nice. This reminds me of that "groups" feature request for HN...

added you as a friend. I'm 12k as well, though I have a lot of variance in my skill level.

I've kept out of KGS for a long time since it was just too tempting, but this post has inspired me to get back into action. I'll also be looking out for other HN readers, username d00d.

The post gives a reasonable overview of why programmers may benefit from playing Go, but there is a whole lot more that could be said. Though I've not yet mastered either pursuit, I'll struggle to put some of my thoughts into words:

Go is particularly suited as an exercise for the programmer's mind because both activities share some quite specific and fundamental cognitive skills, which people generally don't have unless they train.

The overwhelming complexity of Go demands a specific and subtle mindset, which is very much applicable to programming; relaxed yet focused, careful but creative.

Both practise mentally modelling, analysing, and manipulating a complex, evolving system, requiring frequent decisions which may have far reaching and unpredictable effects, which must balance immediate needs and possible outcomes.

Go trains a holistic way of thinking, which is unusual to the average reductionist mind, requiring an integrative understanding of a system as composed of interrelating parts, and as a whole.

Good thoughts - I too have a lot to learn with both Go and programming, and I know there's a lot more going on (with each) than in my original post.

This article is irritatingly full of stretches.

Go proverbs? Are there not chess proverbs? Similar can be said about managing companies and reading Sun Tsu's Art of War. You can extend the Art of War to interpersonal relationships. You can find meaning in every sentence in Harry Potter, if you felt like it.

And Go can make you smarter? Go, like chess, involves short term and long term memory manipulation, risk analysis, pattern recognition, psychology, and executive planning, which further involves deception and short-term and long-term planning. While Go has more possible configurations than Chess, the feeble human mind will not be using some drastically different skillset to deal with it. That is to say, I doubt children who play Chess are worse off than those who play Go. Ergo, it's not Go: it's just the training of relevant skills, which is a byproduct of learning the game.

Pardon my bluntness, but this reads like those articles that try to make something more mystical than it really is.

I think that most of what I said about Go could also apply to chess. I just have less experience with chess than with Go.

I first heard of Go when the film Pi came out: http://en.wikipedia.org/wiki/%CE%A0_(film)#The_game_of_Go

It has conjured up visions of crazy hackers ever since.

Bug in the URL parser?

go is much more interesting than chess. chess is a battle to the death, in go you can have entire games without a single capture, and win by a point or two.

Do you think that the fact that you can have entire games without a single capture, and win by a point or two makes it a more interesting game, or are those two unrelated thoughts?

It's probably a side effect of being a more interesting game, whatever that means to be more interesting, at least.

In chess you could have a checkmate without a single capture. Fool's mate, right?

Few stones being captured is usually a sign of skillful players. I don't think the same can be said of fool's mate.

Anywhere to play GO in Flash or SVG against a computer in different levels?

The only one I found is Yahoo Games but that one sucks and it doesn't have single player.

Why do you want to play against a computer?

Did SVG became Turing complete?

Reply on news.ycombinator.com