Lee Sedol Beats AlphaGo in Game 4

Post by grumpy-buffalo » Sun, Mar 13, 2016, 1:50 PM UTC

denim_chicken wrote: post #190

mikeyouse wrote: post #11

Relevant tweets from Demis; Lee Sedol is playing brilliantly! #AlphaGo thought it was doing well, but got confused on move 87. We are in trouble now... Mistake was on move 79, but #AlphaGo only came to that realisation on around move 87 When I say 'thought' and 'realisation' I just mean the output of #AlphaGo value net. It was around 70% at move 79 and then dived on move 87 Lee Sedol wins game 4!!! Congratulations! H…

It feels really weird to see someone being showered with congratulations for beating a computer program. What exactly is he being congratulated for? For probably triggering and then capitalizing on a bug in AlphaGo's AI? For showing that human resolve, perseverance and a "fighting spirit" can trump a flawed AI, at least until the AI gets fixed? For giving DeepMind extremely valuable test data that will only accelerat…

Post by alphydan » Sun, Mar 13, 2016, 1:54 PM UTC

praptak wrote: post #121

kronion wrote: post #48

After AlphaGo won the first three games, I wondered not if the computer had reached and surpassed human mastery, but instead how many orders of magnitude better it was. Given today's result, it may be only one order, or even less. Perhaps the best human players are relatively close to the maximum skill level for go, and that the pros of the future will not be categorically better than Lee Sedol is today.

Pros themselves estimate their strength 3-4 stones handicap below God: http://senseis.xmp.net/?KamiNoItte :

Post by hyperpape » Sun, Mar 13, 2016, 1:59 PM UTC

keypusher wrote: post #40

The crucial play here seems to have been Lee Seedol's "tesuji" at White 78. From what I understand this phrase in Go means something like "clever play" but is something like sneaking up on your opponent with something that they did not see coming. Deepmind CEO confirmed that the machine actually missed the implications of this move as the calculated win percentage did not shift until later. https://twitter.com/demish…

Post by imron » Sun, Mar 13, 2016, 2:02 PM UTC

jacquesm wrote: post #158

argonaut wrote: post #51

If it's true that AlphaGo started making a series of bad moves after its mistake on move 79, this might tie into a classic problem with agents trained using reinforcement learning, which is that after making an initial mistake (whether by accident or due to noise, etc.), the agent gets taken into a state it's not familiar with, so it makes another mistake, digging an even deeper hole for itself - the mistakes then co…

The same happens to people, especially people that study theory. You can totally throw them off their game by making a non-standard move, even a relatively bad one as long as it breaks their existing pre-conceived notions about how the game should progress. Of course against a really strong player you're going to get beaten after that but a weak player strong on theory will have a harder time.

Post by argonaut » Sun, Mar 13, 2016, 2:11 PM UTC

kotach wrote: post #131

argonaut wrote: post #130

Earlier quoted context omitted.

Yes, I'm pretty sure we're not talking about the same thing. I'm precisely talking about the trajectories not seen problem. Nothing is going to save you from the fact that the net has not seen a certain state before.

That's not really a problem. Given a large enough dataset you want to generalize from it - there are always states not present in the dataset - the whole point is now to extract features out of your dataset to allow generalization on unseen states. Seeing all of the Go games isn't possible. The compounding errors problem that stems from decision bias isn't because you haven't seen the trajectory, it is because the mo…