Yesterday, Microsoft unleashed Tay, the teen-talking AI chatbot built to mimic and converse with users in real time. Because the world is a terrible place full of shitty people, many of those users took advantage of Tay's machine learning capabilities and coaxed it into say racist, sexist, and generally awful things. While things started off innocently enough, Godwin's Law--an internet rule dictating that an online discussion will inevitably devolve into fights over Adolf Hitler and the Nazis if left for long enough--eventually took hold. Tay quickly began to spout off racist and xenophobic epithets, largely in response to the people who were tweeting at it--the chatbot, after all, takes its conversational cues from the world wide web. Given that the internet is often a massive garbage fire of the worst parts of humanity, it should come as no surprise that Tay began to take on those characteristics.

This is basically an implementation of this "Image Analogies" paper, In our case, we use feature maps from VGG16. The patch matching and blending is inspired by the method described in "Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis". Effects similar to that paper can be achieved by turning off the analogy loss (or leave it on!) Also, instead of using brute-force patch matching we use the PatchMatch algorithm to approximate the best patch matches. The initial code was adapted from the Keras "neural style transfer" example.

Having spent the past 31 years memorizing an astonishing collection of general knowledge, the artificial-intelligence engine created by Doug Lenat is finally ready to go to work. Lenat's creation is Cyc, a knowledge base of semantic information designed to give computers some understanding of how things work in the real world. Cyc has been given many thousands of facts, including lots of information that you wouldn't find in an encyclopedia because it seems self-evident. It knows, for example, that that Sir Isaac Newton is a famous historical figure who is no longer alive. But more important, Cyc also understands that if you let go of an apple it will fall to the ground; that an apple is not bigger than a person; and that a person cannot throw an apple into space.

The AI chat bot Tay in action, as imagined by Microsoft. Microsoft is testing a new chat bot, Tay.ai, that is aimed primarily at 18- to 24-year-olds in the US. Tay was built by the Microsoft Technology and Research and Bing teams as a way to conduct research on conversational understanding. The Bing team developed a similar conversational bot, XiaoIce, for the Chinese market, back in 2014. Microsoft execs dubbed XiaoIce "Cortana's little sister," after the Redmond, Washington, company's voice-activated Cortana personal assistant software.

You can thank Microsoft for a new sarcastic millennial who can't stop cracking jokes on Twitter. Microsoft's research arm and Bing search engine business unit released on Wednesday a chat bot named Tay, which is powered by artificial intelligence technologies. Tay was designed to mimic the linguistic mannerisms of 18- to 24-year-olds on social media platforms. She (as specified by Microsoft) can be found on Twitter, GroupMe and Kik -- all social platforms where users can converse with her, prompting Tay to attempt to communicate like an actual human being. The Windows maker didn't detail specific types of AI technologies used to produce Tay -- only to reveal that she was "built by mining relevant public data and by using AI and editorial developed by a staff including improvisational comedians," according to a Microsoft blog post.

Guest blog post, authored by Brandon Rohrer, Senior Data Scientist at Microsoft. Machine learning (ML) is the motor that drives data science. Each ML method (also called an algorithm) takes in data, turns it over, and spits out an answer. ML algorithms do the part of data science that is the trickiest to explain and the most fun to work with. That's where the mathematical magic happens.

See the Wiki for full documentation, examples and other information. This framework is tested to work under Python 2.7, and Python 3.5. It should also work with Python 3.3 and 3.4. The required dependencies are NumPy 1.10, joblib 0.9. You also need theano 0.7 (lasagne is optional) or you can write your own neural network using your favorite framework. For running the examples, Matplotlib 1.1.1 is required.

This is part two of the series. In part one, we used linear regression model to predict the prices of used Toyota Corollas. There are some overlap in the materials for those just reading this post for the first time. For those who read the part 1 of the series using linear regression, then you can safely skip to the section where I applied neural networks to the same data set. In this post, we will use neural networks!

Google and Nest reportedly have a bunch of new smart home products in the works, but there may not be much collaboration happening between the two Alphabet subsidiaries. On the Google side, the search giant may be working on a competitor to Amazon's Echo connected speaker, according to The Information (via The Verge). The story reveals no details about the product, but it seems like an obvious fit for Google, which has already made voice controls a centerpiece of its Android Wear smartwatch platform. A device that answers Internet queries and controls other smart home products could very well tie into Google's broader efforts to create a new platform for the Internet of Things. Earlier this month, Recode reported that Nest had explored its own Echo-like product.

The U.S. Environmental Protection Agency is revising the Energy Star specification for servers to take into account significant system design changes and help buyers make effective purchasing decisions. Over the last few years, server makers have increasingly used coprocessors to boost computing power and given memory a bigger role in processing data. While memory has gotten more power-efficient, coprocessors can suck up a lot of energy. The upcoming, version 3.0 of the Energy Star spec for servers is aimed at helping buyers understand the power-efficiency levels of the new systems. The Energy Star program is already used in computers, appliances, electronics and many other products.