Embedding covert streams into a cover channel is a common approach to circumventing Internet censorship, due to censors' inability to examine encrypted information in otherwise permitted protocols (Skype, HTTPS, etc.). However, recent advances in machine learning (ML) enable detecting a range of anti-censorship systems by learning distinct statistical patterns hidden in traffic flows. Therefore, designing obfuscation solutions able to generate traffic that is statistically similar to innocuous network activity, in order to deceive ML-based classifiers at line speed, is difficult. In this paper, we formulate a practical adversarial attack strategy against flow classifiers as a method for circumventing censorship. Specifically, we cast the problem of finding adversarial flows that will be misclassified as a sequence generation task, which we solve with Amoeba, a novel reinforcement learning algorithm that we design. Amoeba works by interacting with censoring classifiers without any knowledge of their model structure, but by crafting packets and observing the classifiers' decisions, in order to guide the sequence generation process. Our experiments using data collected from two popular anti-censorship systems demonstrate that Amoeba can effectively shape adversarial flows that have on average 94% attack success rate against a range of ML algorithms. In addition, we show that these adversarial flows are robust in different network environments and possess transferability across various ML models, meaning that once trained against one, our agent can subvert other censoring classifiers without retraining.

We apply methods of machine-learning, such as neural networks, manifold learning and image processing, in order to study 2-dimensional amoebae in algebraic geometry and string theory. With the help of embedding manifold projection, we recover complicated conditions obtained from so-called lopsidedness. For certain cases it could even reach $\sim99\%$ accuracy, in particular for the lopsided amoeba of $F_0$ with positive coefficients which we place primary focus. Using weights and biases, we also find good approximations to determine the genus for an amoeba at lower computational cost. In general, the models could easily predict the genus with over $90\%$ accuracies. With similar techniques, we also investigate the membership problem, and image processing of the amoebae directly.

Ynglet is not the first video game to cast its player as an amoeba. Without an expansive repertoire of abilities to call upon, the single cell organism protagonist presents a certain kind of game designer with an alluring challenge. When all a player can do is swim, dash and float about a bit, puzzle and challenge must be carefully constructed from first principles. If it works, as in Ynglet, a meditative, chic jewel of a short game from Swedish designer Nicklas Nygren, there is a purity and clarity of design that none of the medium's sprawling action epics, with their mythically capable warriors, can rival. Contrary to appearance, this is not a game viewed top-down, as if through a microscope's viewfinder.

One of the oldest problems in computer science was just solved by a single cell. A group of researchers from Tokyo's Keio University set out to use an amoeba to solve the Traveling Salesman Problem, a famous problem in computer science. The problem works like this: imagine you're a traveling salesman flying from city to city selling your wares. You're concerned about maximizing your efficiency to make as much money as possible, so you want to find the shortest path that will let you hit every city on your route. There's no simple mathematical formula to find the most efficient route for our salesman.

It is one of the most simple organisms on Earth. But the amoeba, a single-celled organism consisting mostly of gelatinous protoplasm, could be far smarter that thought - and change computing forever. Researchers found they have unique computing abilities that could one day rival conventional computers. The researchers adapted the problem so the amoeba, which can deform its body, was able to use a specially designed chip with 64 'legs', pictured Amoeba, are a single-celled organism consisting mostly of gelatinous protoplasm. The particular type of amoeba that the scientists used was a plasmodium or'true slime mold,' which weighs about 12 mg and consumes oat flakes.

Making a one-legged robot that moves is very hard. Two-legged robots are a little bit more straightforward in some ways, and four-legged robots are statically stable much of the time. You can see where this is going--there's a general trend toward more legs being more stable and potentially easier to control, especially as terrain complexity increases. So what happens if you take that logic to an extreme? We invite you to look at our various contributors @AthisNews - check at the end of our article to discover and follow them on their social networks.

Making a one-legged robot that moves is very hard. Two-legged robots are a little bit more straightforward in some ways, and four-legged robots are statically stable much of the time. You can see where this is going--there's a general trend towards more legs being more stable and potentially easier to control, especially as terrain complexity increases. So what happens if you take that logic to an extreme? As it turns out, you end up with a spherical robot made of 32 individually actuated telescoping legs, named Mochibot.