How an A.I. ‘Cat-And-Mouse Game’ Generates Believable Fake Photos • Hacker News Board

Re: How an A.I. ‘Cat-And-Mouse Game’ Generates Believable Fake Photos

#81

Post by floofyreal » Fri, Jan 05, 2018, 2:30 PM UTC

fjsolwmv wrote: post #44

EGreg wrote: post #26

Mr. Hwang believes the technology will evolve into a kind of A.I. arms race pitting those trying to deceive against those trying to identify the deception. That's like a chess game. We have seen AlphaGo and other MCTS implementations take the "trying to detect the deception" into account. By the time the image is generated, it would have already been factored in.

Where does AlphaGo try to detect deception? What is deception in perfect information games?

https://www.youtube.com/watch?v=XaQu7kkQBPc

Imagine automated system for danger recognition on for example airport. These kind of deception attacks could make problems with these systems. Imagine if suddenly 10,20,100 airports all around globe would recognize weapons, bombs or any other dangerous items? I can imagine panic and huge news headlines badmouthing AI.

People don't trust AI. These kind of errors could only prolong proper integration, which in many ways could enhance the way we live.

Re: How an A.I. ‘Cat-And-Mouse Game’ Generates Believable Fake Photos

#82

Post by bitL » Fri, Jan 05, 2018, 3:26 PM UTC

gwern wrote: post #78

bitL wrote: post #77

Earlier quoted context omitted.

Likely as value function approximators for one piece of the whole algorithm (as is the case with DQN/DDQN). However the main algorithm is likely using variation of Bellman equation, that assumes Markovian property and gives strong guarantees about convergence.

If you're using DQN or pretty much anything in DRL, you don't have any guarantees about convergence in the first place, and using a RNN does give you the history summary you need (at least up to the minimum error achievable with that fixed-length summary, not that that is any more likely to converge than the overall DRL algo is).

I meant that under Markovian assumption value iteration used for Bellman equation is guaranteed to converge. So it makes math people happy, even if such property doesn't hold in the real world nor in the problem they try to solve, and the "deep" in DRL is just heuristics, though surprisingly working in many cases.

Re: How an A.I. ‘Cat-And-Mouse Game’ Generates Believable Fake Photos

#83

Post by amenod » Sat, Jan 06, 2018, 8:53 PM UTC

vinn124 wrote: post #65

amenod wrote: post #39

Earlier quoted context omitted.

> The most famous problem of GANs is instability during training and mode collapse - which is like a student learning especially for an exam (and not in general) thus optimising for the test instead of the real thing. I must confess I haven't worker with GANs yet, but isn't that the whole point of GANs? Student is optimising for the test while the teacher is learning how to make tests as similar to reality as possibl…

> but isn't that the whole point of GANs? not quite, but youre on the right path. think about it this way: you (the generative model) are trying to predict a unit gaussian, which is just a fancy way to say bell curve. you get +1 if you predict a number in this distribution (eg 0.1 or -0.5, which is within one standard deviation of the mean of 0); you get -1 if you predict a number thats "far" from this distribution (…

Thanks!