Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Images altered to trick machine vision can influence humans too

deepmind.google

Post reply on HN
Print view View original

81–86 of 86 posts

Re: Images altered to trick machine vision can influence humans too

#81

Post by hervature »

RandomLensman wrote: post #74
hervature wrote: post #52

Earlier quoted context omitted.

I'm not really sure what you are commenting on. They tested this across multiple pairs of categories: (sheep, chair), (dog, bottle), (cat, truck), and (elephant, clock). This isn't a phenomena related to cats. The whole point of the study is to measure the impact of the noise. The "baseline" or control here would be to to not add noise to either of the two images and arbitrarily label one "cat" and the other "truck"…

1) It is not adding pure noise. 2) If humans when prompted tend to always see something more in one picture than the other when random noise is added, the baseline might not be 50/50 as no matter what you ask you get a systematic preference. Double blinding would not remove this.

Ok, I understand what you are saying. Essentially, you would have liked different type of perturbations tested to get a baseline effect of how much a random perturbation can get people to agree. They did do this in what they call Experiment 3. Their control perturbation is simply the adversarial perturbation but flipped left/right. They claim this was to preserve perturbation statistics. Still, I think baselining against a control perturbation is not the point of the study. My takeaway of the study is that they give a constructive way of influencing human perception of images.

I understand that the concern could be something like "random perturbations of cat images make every image simultaneously less cat-like and hence more like anything else". My opinion is that Experiment 4 (making an image more cat-like or truck-like) covers concerns of this nature. Even if there are two random perturbations where one makes an image cat-like and the other more truck-like, it is completely arbitrary whether you label the perturbation as cat-like or truck-like (since they are randomly generated). That means, even if you measure a difference (and even if two random perturbations have larger differences than this construction!), you cannot control the direction. This method gives you a way to control it. Personally, I don't think this study is about measuring the influence over some baseline. It's about showing that you can indeed choose the direction of influence.

Re: Images altered to trick machine vision can influence humans too

#82

Post by RandomLensman »

hervature wrote: post #81
RandomLensman wrote: post #74

Earlier quoted context omitted.

1) It is not adding pure noise. 2) If humans when prompted tend to always see something more in one picture than the other when random noise is added, the baseline might not be 50/50 as no matter what you ask you get a systematic preference. Double blinding would not remove this.

Ok, I understand what you are saying. Essentially, you would have liked different type of perturbations tested to get a baseline effect of how much a random perturbation can get people to agree. They did do this in what they call Experiment 3. Their control perturbation is simply the adversarial perturbation but flipped left/right. They claim this was to preserve perturbation statistics. Still, I think baselining aga…

I agree that in some ways they control for certain aspects, but with human behavior I am leaning towards being old school in wanting to see baselines when there is nothing "structured" to have a better idea of baselines (and potentially also learn something about human perception). In the current study I am not fully convinced all confounders are controlled enough. (There is also the issue of prompting, but asking if it were not, e.g., a vase what else is there is experimentally too broad, I fear)

A broader way could be to add random noise then ask to pick the more X-like imagine and see how that correlates with the classifier probability for X.

Re: Images altered to trick machine vision can influence humans too

#83

Post by EnigmaFlare »

krisoft wrote: post #70
EnigmaFlare wrote: post #67

It would be amazing if NNs can eventually create art that inspires emotions in the audience without them realizing how. That seems to be a feature of normal art but maybe machines can one day do better than the best human artists. Unfortunately, the researchers will probably never follow up on their work and improve things. That's the sad story of most fun research, they just give up and forget about it once it's pub…

> Unfortunately, the researchers will probably never follow up on their work That is a very weird assumption. > nobody else seems to want to do the work themselves If you are so interested why don’t you try your hand at it? This particular research is super easy and accessible.

That's fair. Of course they're not obligated to continue it themselves. I guess I'm just speaking from disappointment when I've encountered old, seemingly interesting dead-ends before and wondered why the authors apparently abandoned their work when they'd be the ones most motivated to further it. Perhaps getting the ball rolling is more the researcher's job than improving the ball.

Re: Images altered to trick machine vision can influence humans too

#84

Post by refulgentis »

Terr_ wrote: post #63
ndriscoll wrote: post #62

Earlier quoted context omitted.

I don't think it is subconscious. In the bottom left of the vase picture, the "cat-like" version has "whiskers". The egg has a sort of caty blob on the right (though it's hard for me to decide whether the cat or truck version is more pronounced). The street picture has a white blob in the center that's "sheepier" than the chair version. To me it seems like a conscious "if I had to pick the more cat-like one, this pic…

In the case of giving people to look at subtly-distorted images and asking them "truck-vs-cat" and "chair-vs-sheep", I think it's likely people thought: "Huh, I guess one of those involves artificially constructed angles and the other tends to organic rounded bits, so I'll decide on which seems extra-present." To me that seems a reasonable--and not very exciting--explanation for why humans may get similar this-or-tha…

Yeah but that's not what happened

Re: Images altered to trick machine vision can influence humans too

#85

Post by krisoft »

glitchc wrote: post #79
krisoft wrote: post #71

Earlier quoted context omitted.

> It requires asking an open-ended question such as "what does this image look like to you?" Because the answer to that is simply “it looks like flovers in a vase”. There is no question about human’s ability to tell what the image is. So much so that if you ask the humans to describe the images they would probably say something along the lines of “two identical images of the same flowers”. So you would think if you a…

> Because the answer to that is simply “it looks like flovers in a vase”. There is no question about human’s ability to tell what the image is. Thank you. Because that right there means that any bias being measured is one that's introduced by the researchers. Ergo the study is useless.

> bias being measured is one that's introduced by the researchers

Yes. Intentionally.

> Ergo the study is useless.

No, it is not useless. It just not studying what you seem to think it does.

There was a previous very well estabilished finding that you can manipulate images such that they look basically identical to humans but they completely change classification for the neural network. This is an estabilished fact.

Have you heard about this? Are you aware of this? Because if you are not that would explain your confusion.

This research is not proving that phenomenon. It has been proven before. It follows up on it and investigates what that “basically identical” means.

There are two assumptions one can make:

1; The difference between the manipulated and the not manipulated image is so small humans won’t be able to tell which one is wich.

2; Since the manipulations appear to be neural network specific there is no reason to expect that a human will preceive them as the same class as the neural network. For all we know the humans might see the original image as more cat like, or the one we manipulated to look like a truck will appear more cat like to a human.

These are the two statements they wanted to investigate.

Re: Images altered to trick machine vision can influence humans too

#86

Post by im3w1l »

wizzwizz4 wrote: post #80
hytfyiv3j wrote: post #65

Earlier quoted context omitted.

If you actually went through with it, traced over that picture and posted it back to the thread that would be nice, as it seems not everyone can see it as clearly as you can. Maybe even the original authors would appreciate it.

https://0x0.st/H62J.ora This is not a permanent link, and it's not my server so please don't hammer it. Well, it turns out I can't mouse-draw, even when I'm tracing, but I've given it a go. It's supposed to be a cat walking from the right to the left of the image. The first layer has a better head, and the second layer has a better front leg and ears. Note that my lines obscure the recognisable features, so you'll ha…

I took a look and my brain must be wired differently from yours because I don't really see it.
Post reply on HN