Wow, awesome work tj86430, thank you very much! I'm new to this forum, I didn't know we can include a web based Go client here

Ah! What about trying it with Black? First move pass, then apply the same strategy.

In case you were wondering Thomas, my post wasn't actually directed at you. I figure at 1d this is likely an exercise in amusement as opposed to a serious HaXX0r win strategy for those hardcore bot-players out there

However, I've been looking at sensei's articles recently on "ripoffs that GnuGo answers wrong", and I really think people get carried away with the result a bit too much. This to me is a fabulous bit of research for GnuGo engine writers to fix, and it's a good laugh. My concern is that weaker players will read it, and brag about how they can beat an SDK every time, and wonder why their strategy doesn't work against real players.

This sort of thing makes me want to return to my idea of writing heuristic Go playing engines :S

My client hangs when I pass on my first turn with black, so I played Ten Gen instead. Here's the outcome:

[sgf-full](;FF[1]GM[1]SZ[19]AP[Jago:Version 4.53]GN[GloFish vs GnuGo 4];B[jj];W[dp];B[eo];W[do];B[en];W[cm];B[dl];W[fq];B[gp];W[dd];B[ee];W[cf];B[dg];W[fc];B[gd];W[pd];B[oe];W[nc];B[md];W[qf];B[pg];W[pp];B[oo];W[nq];B[mp];W[qn];B[pm];W[gq];B[hp];W[ir];B[jq];W[rl];B[qk];W[mc];B[ld];W[kb];B[jc];W[rh];B[qi];W[bh];B[ci];W[pn];B[on];W[cl];B[dk];W[gc];B[hd];W[np];B[mo];W[lr];B[mq];W[jr];B[kq];W[bi];B[cj];W[jb];B[ic];W[ib];B[lc];W[no];B[nn];W[mr];B[op];W[oq];B[po];W[qo];B[qm];W[rp];B[od];W[oc];B[pe];W[qe];B[ed];W[ec];B[de];W[ce];B[cg];W[bg];B[dn];W[cn];B[ep];W[eq];B[hq];W[rm];B[hr];W[pq];B[lq];W[hs];B[gr];W[dr];B[qg];W[rg];B[ql];W[kr];B[hc];W[lb];B[hb];W[dc];B[df];W[qd];B[pf];W[bj];B[ck];W[bk];B[fp];W[ha];B[gb];W[fb];B[ga];W[ri];B[qh];W[rk];B[rj];W[sj];B[qj];W[gs];B[fr];W[er];B[fs];W[is];B[es];W[ds];B[dm];W[ia];B[fa];W[ea];B[kc];W[fd];B[fe];W[db];B[ch];W[nd];B[ne];W[iq];B[ip];C[Passen];C[Passen];C[Passen]TW[aa][ab][ac][ad][ae][af][ag][ah][ai][aj][ak][al][am][an][ao][ap][aq][ar][as][ba][bb][bc][bd][be][bf][bl][bm][bn][bo][bp][bq][br][bs][ca][cb][cc][cd][co][cp][cq][cr][cs][da][dq][eb][ja][js][ka][ks][la][ls][ma][mb][ms][na][nb][nr][ns][oa][ob][or][os][pa][pb][pc][pr][ps][qa][qb][qc][qp][qq][qr][qs][ra][rb][rc][rd][re][rf][rn][ro][rq][rr][rs][sa][sb][sc][sd][se][sf][sg][sh][si][sk][sl][sm][sn][so][sp][sq][sr][ss]TB[dh][di][dj][ef][eg][eh][ei][ej][ek][el][em][ff][fg][fh][fi][fj][fk][fl][fm][fn][fo][ge][gf][gg][gh][gi][gj][gk][gl][gm][gn][go][he][hf][hg][hh][hi][hj][hk][hl][hm][hn][ho][id][ie][if][ig][ih][ii][ij][ik][il][im][in][io][jd][je][jf][jg][jh][ji][jk][jl][jm][jn][jo][jp][kd][ke][kf][kg][kh][ki][kj][kk][kl][km][kn][ko][kp][le][lf][lg][lh][li][lj][lk][ll][lm][ln][lo][lp][me][mf][mg][mh][mi][mj][mk][ml][mm][mn][nf][ng][nh][ni][nj][nk][nl][nm][of][og][oh][oi][oj][ok][ol][om][ph][pi][pj][pk][pl])[/[/sgf-full]

It's just one game so far, but it seems to be pretty much the same as before. I guess, GnuGo is even more discouraged to play an invasion now, due to the Ten Gen stone. Since I never had to resort to the komi points to win with white, I don't see a reason why it shouldn't work the same way when playing black.

So a programming mistake forces you to lose 1 point.

Your games also show that the center is bigger than corners and edges if defended only. Why would I have expected this?:)

Glofish, you truly deserve the batman smiley. These games are hilarious. Thanks for exposing GnuGo as the bot that it is.

_________________
Patience, grasshopper.

I thought this was a lame thread until I saw the games. That's a pretty large flaw you've found there! Very amusing.

_________________
That which can be destroyed by the truth should be.
--
My (sadly neglected, but not forgotten) project: http://dailyjoseki.com

I'm not so sure. Not so sure that this is a specific flaw rather than a clear example of a more general problem. That the "evaluator" component (that chooses the next move) is unable to take into account global considerations, decides based just upon what it "thinks" is locally the best move even though that might be bad in global terms.

But isn't this a more or less typical failure at this level of playing strength? Or to put it another way, why gnugo's strength might be only 8k (or whatever it is) even though its local play strength might be higher than that? Don't human 8k's also fail to maintain "balance" between their positions in different parts of the board. Don't they play what they have learned as the best joseki irregardless of the fact that given what has been played in the other corners, they should play differently here.

Does a game between a human 1d and a human 8k (or whatever) look that different? Doesn't the weaker human player make precisely this sort of mistake in "direction of play"?

Something for the gnugo development team to address of course. Not maintaining balance in fuseki. I think that is the actual problem. Probably could come up with other examples where the program could be "tricked" into choosing local plays that were collectively bad besides the way done in these examples.

Not that way. Even a 20k would have enough fighting spirit to at least try.


Yes. Yes, we kuy players make tons of mistake, but I never saw anyone play so passively.

Are you suggesting that a human player (playing at about an 8 stone disadvantage in strength) would learn that "mixing it up" pays? I seriosuly doubt that. Playing agrressively, leaving behind groups that should/could survive in fights would give them the experience that "all my groups die when I play that way". Aji that is only a minor problem against equal strength opposition is disasterous when the opponent is so much stronger than we are.

The human 8k (or 20k) learns to prepare to break through against opposition not that much stronger than they are and so there are defects allowing them to break through and in the fighting and equal chance that the oppponent will make a mistake or misue the aji left behind. We learn from what succeeds, not from what fails. We don't/won't learn to play properly if our only game experience is playing even games against players that much stronger than ourselves doing their utmost against us

I think we are mistaking what we are seeing here. This "passivity" (failure to analyze globally) is why the program has the rating that it does. This part of its game is weak relative to other parts of its game so in practice the program comes out at about 8k overall (or whatever it is). We need to look at examples of human 8k's trying this "easy win" strategy against gnugo 3.8 . We may find that enough errors will be made in the josekis and sealing off and use of aji that the outcome is different. Conversely we may find that gnugo has a lower effective rating. But looking at games where the human player has about an 8 stone edge tells us little.

NOTE: The MCTS programs wouldn't play so passively but they would make what look like ridiculously bad moves if you force them to play at this disadvantage and don't let them resign. We need to accept that perhaps the computer programs are more sensitive to "the state of the game" than human players would be.

Sorry, I don't feel the need to continue the discussion. Try to find some real games with meat&bones players. There are 8 stones difference games played everyday (check out the ASR League for instance), go look for a few of them.

I'm 6k and I just beat gnugo following the above strategy and otherwise playing mindlessly. I am sure that it would take less than 1 hour to teach a 16k to beat it.

_________________
Patience, grasshopper.

Cool, thank you for your report. I'm looking forward to reports of even weaker players, how they could deal with that strategy. My statement that 1 digit kyu's can apply the strategy was only a guess after all, I wonder if it holds.



Sorry Mike, but I think you're totally wrong here. From even games against much stronger players you learn most, cause you can see what moves your opponent applies to crush you, and then try them in the next games. And you also learn from your failures. When a move doesn't work, you learn that it doesn't work and try another one next time, until you found a move that works. And once again, you learn most effectively when playing against a much stronger player, cause you won't be left with any illusions. If your move doesn't work, the stronger player will punish it and thus support your learning. I'm well aware that this kind of learning might be demotivating for some people - nevertheless it is the most effective way to become stronger. I've always challenged stronger players at Go congresses as often as I could - not to win, but to see Go games at high level and learn from them - and I was well rewarded!

Yeah, I'll second those who say an 8k human would not play like this. They might make some of the same initial mistakes, but they would invade.

I'm curious to see how weak a player would need to be before they couldn't apply this strategy. I'm not so sure a 16k wouldn't flub it--just push a little too much, end up with too few liberties, and you're in trouble.

_________________
Occupy Babel!

Not reading what I wrote?

Unless the much stronger player is playing a "teaching game" with you, won't learn that stuff. I was not suggesting that one didn't learn best by playing against a much stronger player. I agree entirely that this disparity in strength will quickly let you see what you are doing wrong, immeidately punish mistakes. It's the "even game" part that I am disagreeing with. To learn fastest what you want is for that much weaker player to have a slightly inadequate handicap.

In other words, I am saying that an 8k won't learn that that much playing a 1d even but would if playing with a 5-6 stone handicap.

An "even" game there would technically be a 8 stone handicap, no? But regardless. I've improved a lot while playing even games with players 5-9 stones stronger than I was. Reviewing even games against stronger players makes your mistakes much more stark (because up until a certain point, you were doing better).

But that is not how the term "even game" has been used in this discussion (remember the examples?)

Jts, you meant "even" in the different sense (with handicap) and I couldn't agree with you more. Playing even games (no handicap) against players as weak as you are will let you get away with too many mistakes. Playing with the game slightly uneven (slightly inadequate handicap) will highlight what you are doing wrong. Frankly, it's no always necessary to go over the game to see. At least in my case usually immediately obvious "darn; I needed to make a defensive move and that was a defensive move that retained sente because with that attacking group safe the attack (that otherwise falls to a counter attack) would win through" --- I give that example because missing these is a current problem of mine.

But I don't think 8-9 stone games best. An opponent several stones stonger will be enough stronger and with the initial board more open (fewer handicap stones) the game closer to an "even game" (no handicap). Of course you can learn from mistakes you are making that an opponent 8-9 ranks higher can spot and punish but that might succeed against an opponent just 4-5 ranks higher. But I say "walk before you run". It's the mistakes typical of your current level that a player 4-5 ranks higher doesn't make you should learn first. Wait till later to learn the things you are doing wrong that a player 4-5 ranks stronger also does wrong.

This being the "computer go" section I think this is very important for those of us who have* to use the computer as an opponent. Just my opinion, but the "bad habits" problem should be far less of a problem if the computer is "peculiar" at its own level of play. Those far less likely to manifest if the playing strength of the program is adjusted to 3-6 stones stronger than you are.




* I get to go to the go club once a week and play against humans there and this time of year that's "weather permitting" (snow either the day the club meets or the day before and too sore/tired after digging out to drive any distance and then stay out late) I can play a game or two a day against the machine with a program enough stronger than I am so I can learn from it.

Sorry, I (stupidly) switched from one sense of even to another in two consecutive sentences. I play no-handicap, 6.5 komi games with stronger players as part of ASR, and I've been learning a lot from it.

A bit of research (gnugo developers correct me if I am wrong)

Since no users of gnogo responded to my semi questions I went to their site to see what I could determine.

a) Deterministic only --- I could find nothing referring to a mode where probablity is involved with move selection. That means gnugo will always respond exactly the same way in the same situation and a weaker player could "learn a game" that defeats it. In other words, it represents a more general weakness than the specific "bad game" example that started this thread.

b) Center influence ----- This is adjustable, at least selectable even if off/on vs a continuously adjustable parameter. But hasn't been made available as a run time parameter so you need to be able to recompile the program ot turn on "cosmic".
The site says that with this on beats gnugo with this off but weaker against human players. That would contradict what has been suggested here that a weaker player could easily "learn" to use gnugo's passivity. We really need to here from players weaker than gnugo is described to be that they are able to do so. Sorry, but that stronger players can do it means nothing as they should defeat gnugo in any case.

Were I on the development team I would ..........
1) Make the "cosmic" option available as a runtime option. Even if off/on. Could be even more interesting to make the "off" vs "on" a matter of probablility (parameter changes the odds whether gnugo will play this game in an "territorial" or "influence" style.

2) Introduce probablity in move selection. Bad to weaken the evaluator too much so I would leave it deterministic if the top move evaluates more than the second best by a certain amount, but this can be almost automatic if the choice between the two top moves is proportional to their metrics. This would make it much harder for a weaker human opponent to "learn a game that defeats gnugo all the time".

However those two suggestion would be only if continuing not to use MCTS as the "evaluator". As things stand now there are two parts to a program like gnugo. One is the AI that has processes that find plausible moves (moves that might be good for some go reason) and the other is an AI that is the "evaluator", tries to determine which of these is the best/most important move at the moment given the current state of the game. But MCTS can be used for the "evaluator" instead.

Huh? The example that started this thread was not a specific "bad game," but rather pointed exactly to a "general weakness."



While you go on to make some of your own valid points, particularly about how GnuGo works and how to improve it, you seem to be missing the point of this thread, which is that a simple mechanical strategy tricks GnuGo into playing a losing strategy. It is irrelevant that a stronger player would win anyway. When I tested the OP's strategy, I did not attempt to win, I simply attempted to employ the theory, and GnuGo played exactly as the OP predicted. That's the point.

_________________
Patience, grasshopper.