Given the well-known limitations of the Turing Test, there is a need for objective tests to both focus attention on, and measure progress towards, the goals of AI. In this paper we argue that machine performance on standardized tests should be a key component of any new measure of AI, because attaining a high level of performance requires solving significant AI problems involving language understanding and world modeling - critical skills for any machine that lays claim to intelligence. In addition, standardized tests have all the basic requirements of a practical test: they are accessible, easily comprehensible, clearly measurable, and offer a graduated progression from simple tasks to those requiring deep understanding of the world. Here we propose this task as a challenge problem for the community, summarize our state-of-the-art results on math and science tests, and provide supporting datasets

You may think you do, but a recent experiment by a Russian photographer suggests otherwise. In a project entitled, "Your face is big data," Rodchenko Art School student Egor Tsvetkov began by photographing about 100 people who happened to sit across from him on the subway at some point. He then used FindFace, a facial-recognition app that taps neural-network technology, to try to track them down on Russian social media site VK. It was ridiculously easy to find 60 to 70 percent of the subjects aged between 18 and 35 or so, he found, although for older people it was more difficult. Along the way, he learned a whole lot about the lives of complete strangers.

I'd seen everything that Grindr had to offer and was growing weary of unsolicited dick pics and random old balls. So I turned to Tinder and had a couple failed dates. The first was a disgruntled state worker who wore Tom's, winced when I told him I had two pitbulls and spent the better part of two hours mansplaining ethics to me. The next was a waifish first-year English teacher and self-professed INFJ who, curiously, didn't do a lot of reading due to his workload. I powered through, reminded of months of success, but couldn't help thinking of my failure.

Roughly a year and a half ago, I had the privelage of taking a graduate "Introduction to Machine Learning" course under the tutelage of the fantastic Professor Leslie Kaelbling. While I learned a great deal over the course of the semester, there was one minor point that she made to the class which stuck with me more than I expected it to at the time: before using a really fancy or sophisticated or "in-vogue" machine learning algorithm to solve your problem, try a simple Nearest Neighbor Search first. Let's say I gave you a bunch of data points, each with a location in space and a value, and then asked you to predict the value of a new point in space. Perhaps the values of you data are binary (just s and -s) and you've heard of Support Vector Machines. Should you give that a shot?

Is there a reason these are not employed more often in Machine Learning problems and more often in scientific computing?

Imagine a world without government, schools, a legal system, law enforcement, or companies. It's a world unlike the one we currently live in--but based on the evolution of technology and how we use it--representative of what the world may become. Imagine a computer infrastructure that could--with global knowledge and the ability to enact precise tweaks to the social and economic structure--drive the evolution of society. This is the idea behind the Universal Graph. In mathematics, this is a graph (or network) in which a piece of information can be connected with other pieces of information until all finite information is integrated.

We propose a new model for learning bilingual word representations from non-parallel document-aligned data. Following the recent advances in word representation learning, our model learns dense real-valued word vectors, that is, bilingual word embeddings (BWEs). Unlike prior work on inducing BWEs which heavily relied on parallel sentence-aligned corpora and/or readily available translation resources such as dictionaries, the article reveals that BWEs may be learned solely on the basis of document-aligned comparable data without any additional lexical resources nor syntactic information. We present a comparison of our approach with previous state-of-the-art models for learning bilingual word representations from comparable data that rely on the framework of multilingual probabilistic topic modeling (MuPTM), as well as with distributional local context-counting models. We demonstrate the utility of the induced BWEs in two semantic tasks: (1) bilingual lexicon extraction, (2) suggesting word translations in context for polysemous words. Our simple yet effective BWE-based models significantly outperform the MuPTM-based and context-counting representation models from comparable data as well as prior BWE-based models, and acquire the best reported results on both tasks for all three tested language pairs.

A key front for ethical questions in artificial intelligence, and computer science more generally, is teaching students how to engage with the questions they will face in their professional careers based on the tools and technologies we teach them.  In past work (and current teaching) we have advocated for the use of science fiction as an appropriate tool which enables AI researchers to engage students and the public on the current state and potential impacts of AI. We present teaching suggestions for E.M. Forster's 1909 story, "The Machine Stops," to teach topics in computer ethics.  In particular, we use the story to examine ethical issues related to being constantly available for remote contact, physically isolated, and dependent on a machine --- all without mentioning computer games or other media to which students have strong emotional associations. We give a high-level view of common ethical theories and indicate how they inform the questions raised by the story and afford a structure for thinking about how to address them.

Influence Maximization is an active topic, but it was always assumed full knowledge of the social network graph. However, the graph may actually be unknown beforehand. For example, when selecting a subset of a homeless population to attend interventions concerning health, we deal with a network that is not fully known. Hence, we introduce the novel problem of simultaneously influencing and mapping (i.e., learning) the graph. We study a class of algorithms, where we show that: (i) traditional algorithms may have arbitrarily low performance; (ii) we can effectively influence and map when the independence of objectives hypothesis holds; (iii) when it does not hold, the upper bound for the influence loss converges to 0. We run extensive experiments over four real-life social networks, where we study two alternative models, and obtain significantly better results in both than traditional approaches.