Slashdot Mirror
Research Highlights How AI Sees and How It Knows What It's Looking At
anguyen8 writes Deep neural networks (DNNs) trained with Deep Learning have recently produced mind-blowing results in a variety of pattern-recognition tasks, most notably speech recognition, language translation, and recognizing objects in images, where they now perform at near-human levels. But do they see the same way we do? Nope. Researchers recently found that it is easy to produce images that are completely unrecognizable to humans, but that DNNs classify with near-certainty as everyday objects. For example, DNNs look at TV static and declare with 99.99% confidence it is a school bus. An evolutionary algorithm produced the synthetic images by generating pictures and selecting for those that a DNN believed to be an object (i.e. "survival of the school-bus-iest"). The resulting computer-generated images look like modern, abstract art. The pictures also help reveal what DNNs learn to care about when recognizing objects (e.g. a school bus is alternating yellow and black lines, but does not need to have a windshield or wheels), shedding light into the inner workings of these DNN black boxes.
Unfortunately they are wrapped around a tree; just around the corner. Mistook a bee 3 inches from the camera for a school bus.
John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
Reminds me of the reverse OCR tumblr. It generates patterns of squiggles a human could never read but the OCR recognizes as a word.
http://reverseocr.tumblr.com/
idk, these results seem more similar to how humans see than they do different. When people don't know exactly what they are looking at, the brain just puts in it's best guess. people certainly see faces and other familiar objects in tv static. They see bigfoot in a collection of shadows or a strange angle on a bear. i even feel like i did sort of see a peacock in the one random image labeled peacock. it's sort of like the computer vision version of a rorschach test.
> It's OK, if AI is this stupid, we need not worry about it taking over any time soon.
If only that worked for congress.
"I opened my eyes, and everything went dark again"
If the network was trained to always return a "best match" then it's working correctly. To return "no image", it would need to be trained to be able to return that, just like humans are given feedback when there is no image.
So it needs to learn that these exact images are tricks being played on it, so it can safely ignore it. This is exactly what machine learning is. What's the story?
But don't worry. I'm sure the armchair experts of Slashdot will be along any minute to tell us how this all just a bunch of hype and that the computers are stupid (I'm not disagreeing - for the moment) and AI is at least ten millions years away and will likely NEVER come to pass. Seriously though, I think a large portion of this site's users have their heads in the sand. I don't work in the field, but I am very interested in it and I read a lot of material from a lot of reputable sources. It seems to me that there are some very deep pockets out there treating this as a serious project and are determined to succeed. Personally, I think they will succeed and far sooner than almost everyone will expect. To have a huge impact, AI doesn't have to be perfect. It doesn't have to reason at a human level to be of use or have a noticeable effect on the economy. And once simpler forms of AI arrive, it will advance very rapidly. I think the folks here on Slashdot will be denying the possibility of such a thing right up until the day before they find themselves on the unemployment line. I think we (and our political leaders) should be preparing for a new economy today while there's still time. Otherwise, it'll be a catastrophe for the majority of working people and society at large.
"War makes me sad." - Me
I've done some image processing work.. It seems to me that you can take the output of this Neural network and correlate it with some other image processing routines, like feature detection, feature meteorology, etc; A conditional probability based decision chain,etc.
I work on a LIDAR sensor meant for Anti-. I work at a start-up that makes 3D laser-radar vision sensors for robotics and autonomous vehicles /anti-collision avoidance. The other day, I learned that such sensors allow robots to augment their camera vision systems to have a better understanding of their environment. It turns out that it's still an unsolved problem for a computer vision systems to unambiguously recognize that it's looking at a bird or a cat, and can only give you probabilities.. A LIDAR sensor instantly gives you a depth measurement out to several hundred meters that you can correlate your images to . The computer can combine the color information, along with depth information to have a much better idea of what it's looking at. For an anti-collision avoidance system, it has to be certain what it's looking at, and that cameras alone aren't good enough. I find it pretty exciting to be working on something that is useful for AI (artificial intelligence) research. One guy I work with got his Ph.D using Microsoft's Kinect sensor, which is something that gives robots depth perception for close-up environments..
“In the 60s, Marvin Minsky (a well known AI researcher from MIT, whom Isaac Asimov considered one of the smartest people he ever met) assigned a couple of undergrads to spend the summer programming a computer to use a camera to identify objects in a scene. He figured they'd have the problem solved by the end of the summer. Half a century later, we're still working on it.”
http://imgs.xkcd.com/comics/ta...
There's also a tremendous gap between what we consider complex and what we consider simple. For example, the brain is complex. However, individual elements of our brains are incredibly simple. Basic chemical reactions. Neurons firing or not. It's the sheer number of simultaneous simple pieces working together that makes it complex.
Lots of simple AI algorithms all working together make the complexity. This isn't climbing a tree. It's one person poking at chemicals until they get high-energy combustible fuels, and another playing with paper to make paper airplanes better, and a third refining ceramics and metals to make them lighter and stronger and to handle different characteristics, and then they all get put together and you have a person on the moon.
The illusion is that you think we need to make a leap to get from here to there. There's never a leap. It's lots of small simple steps that get you there.
~Anguirel (lit. Living Star-Iron)
QA: The art of telling someone that their baby is ugly without getting punched.
The DNN examples were apparently trained to discriminate between a members of a labeled set. This only works when you have already cleaned up the input stream (a priori) and guarantee that the image must be an example of one of the classes.
These classifiers were not trained on samples from outside the target set. This causes a forced choice: given this random dot image, which of the classes have the highest confidence? Iterate until confidence is sufficiently high, and you have a forgery with the same features the classifier is looking for.
For example, the digit training set (0,1,2...9) would need to be augmented with pictures of 'A', 'D', a smiley face, a doodle of a tree, a silhouette of Alfred Hitchcock and some spider webs. The resulting classifier would be more robust. The target classes (0,1,2,...9) would be counterbalanced with a null class (everything else). Looking inside the receptive fields of a robust image classifier is rather satisfying: you will find eigenimages that project back to image structures that are human recognizable, too.
The lesson in training your classifier is to either verify your assumption (all incoming samples must be a member of the chosen classes) or train (expose) your classifier to out-of-class samples.