Synthesizing Robust Adversarial Examples

to such sycophancy or boasting. > The whole lobbying apparatus is about altering the perceptions of lawmakers so that you get more favorable laws. People usually operate on fewer pieces of data than computers, and so one falsehood can alter their decision-making more than it would a computers. Yup, absolutely. But if the software developer misses something that an outsider can learn about — this can be something that just didn't even exist at the time the developer made their choices, or just a bug they didn't spot — it can be abused in the same way. When it comes to accidents, the USA almost started WW3 because someone forgot to account for the fact that the moon rises in the same part of the sky that a Soviet strike group would come from, and also doesn't respond to IFF pings. The Soviet union made their own mistake with sun glare. When it comes to things outsiders learn and the original developer didn't think of, other than all bugs and exploits, consider also things like this: https://arxiv.org/pdf/1707.07397

groundtruth categories to allow evaluation. That way, good results clearly prove a good unsupervised training method. The problem that I mean in the second part was that most networks trained for object recognition work on low-level features such as colors and textures, as shown in [3]. The turtle clearly has a turtle shape and arrangement and looks overwhelmingly like a turtle to humans. But its high-frequency surface details are those that the neural network associates with a rifle, which is why those networks are fooled even on photos from varying perspectives. Training a network with a loss to ensure that the local area of an image produces features that are highly correlated to the global features of the same image does not avoid this problem, because the high-frequency patters that the AI erroneously uses for detection are present both in the local as well as in the global scale. Sadly, I don't have any idea on how to improve that either. [1] What's Hidden in a Randomly Weighted Neural Network? https://arxiv.org/abs/1911.13299 [2] Invariant Information Clustering for Unsupervised Image Classification and Segmentation https://arxiv.org/abs/1807.06653 [3] Synthesizing Robust Adversarial Examples https://arxiv.org/abs/1707.07397

[1]. Tacotron2 produced completely passible TTS [2] GANs improved dramatically and saw some good theory to back it up [3] [4]. We also started to care more about how well our models do in general, which to me shows maturity: Adversarial examples showed their teeth, and hopefully convinced everyone to care about robust models [5] [6]. Reinforcement learning algorithms were shown to have poor transferability [7]. In 2018 I hope to see a new kind of CNN. Residual style networks are the norm now, since they mostly solve problems with gradient flow. But take away all the skip connections and we're mostly left with a vgg-style linear net with box filters. I'd be really excited to see a network with image-sized conv filters that could adapt their shape (and therefore representational power) to a given feature or signal. Hopefully 2018 is the year where people stop calling AI a one-trick-pony. I don't hear it as often these days, but I think its time to put that phrase to rest. [1] https://arxiv.org/abs/1712.01815 [2] https://research.googleblog.com/2017/12/tacotron-2-generatin... [3] https://arxiv.org/abs/1710.10196 [4] https://arxiv.org/abs/1701.04862 [5] https://arxiv.org/abs/1707.07397 [6] https://arxiv.org/abs/1710.06081 [7] https://arxiv.org/abs/1709.06560