The European Commission has announced the launch of a pilot project intended to test draft ethical rules for developing and applying artificial intelligence technologies to ensure they can be implemented in practice. It's also aiming to garner feedback and encourage international consensus building for what it dubs "human-centric AI" -- targeting among other talking shops the forthcoming G7 and G20 meetings for increasing discussion on the topic. The Commission's High Level Group on AI -- a body comprised of 52 experts from across industry, academia and civic society announced last summer -- published their draft ethics guidelines for trustworthy AI in December. A revised version of the document was submitted to the Commission in March. It's boiled the expert consultancy down to a set of seven "key requirements" for trustworthy AI, i.e. in addition to machine learning technologies needing to respect existing laws and regulations -- namely: The next stage of the Commission's strategy to foster ethical AI is to see how the draft guidelines operate in a large-scale pilot with a wide range of stakeholders, including international organizations and companies from outside the bloc itself.

Unfortunately for our new AI overlords, the crusade to take over the world has been stopped in its tracks by an unlikely hurdle: a 16-year-old's math test. Faced with the same level of exam that a 16-year-old in the U.K. would take, according to a new paper by Google's DeepMind, its cutting-edge AI flunked. The algorithm was trained on the sorts of algebra, calculus, and other types of math questions that would appear on a 16-year-old's math exam according to the U.K. national curriculum, according to DeepMind research published online on Tuesday. The researchers tested several types of AI and found that algorithms struggle to translate a question as it appears on a test, full of words and symbols and functions, into the actual operations needed to solve it, according to an article on Medium. It turns out, according to the research, that even a simple math problem involves a great deal of brainpower, as people learn to automatically learn to make sense of mathematical operations, memorize the order in which to perform them, and know how to turn word problems into equations.

Sometime in the early 2000s, while sitting in my dentist's chair, I began to wonder about the potential real-world pain that someone could potentially inflict on another human being simply by hacking the new digital x-ray system that the dentist had installed. Would it be possible, for example, for a hacker to modify the digital images from the x-rays so that the dentist would not be able to find and repair painful cavities, or to cause the dentist to perform an unnecessary root canal, filling, or other procedure? How certain could I be that the images of my own teeth were not tampered with? Several years later, when I had my a digital MRI after an auto accident, I wondered even further – could hackers modify images in such a manner so as to cause a person to have his head cut open to remove a tumor when, in fact, he had no tumors? Or to cause a scan to appear normal when the victim actually had a life threatening condition requiring immediate attention?

Last week's breaking news story on The Robot Report was unfortunately the demise of Helen Greiner's company, CyPhy Works (d/b/a Aria Insights). The high-flying startup raised close to $40 million since its creation in 2008, making it the second business founded by an iRobot alum that has shuttered within five months. While it is not immediately clear why the tethered-drone company went bust, it does raise important questions about the long-term market opportunities for leashed robots. The tether concept is not exclusive to Greiner's company, there are a handful of drone companies that vie for marketshare, including: FotoKite, Elistair, and HoverFly. The primary driver towards chaining an Unmanned Ariel Vehicle (UAV) is bypassing the Federal Aviation Administration's (FAA) ban on beyond line of sight operations.

Automation will increasingly allow vehicles to take over certain aspects of driving. However automated functions are still being fine-tuned, for example, to ensure smooth transitions when switching between the human driver and driverless mode. Standards also need to be set across different car manufacturers, which is one of the goals of a project called L3Pilot. Although each brand can maintain some unique features, automated functions that help with navigating traffic jams, parking and motorway and urban driving must be programmed to do the same thing. 'It's like if you rent a car today, your expectation is that it has a gear shift, it has pedals, it has a steering wheel and so on,' said project coordinator Aria Etemad from Volkswagen Group Research in Wolfsburg, Germany.

Video games are one of the most misunderstood forms of entertainment. In one sense, it's easy to see why: if you haven't had much interaction with them, watching someone play one can be a pretty unsettling experience. Gamers can often give the impression that they're glued to the screen, absorbed in what feels like the digital equivalent of junk food. At best, it seems like a pointless thing to do; at worst, we worry that games are socially isolating, or actively harmful. One of the longest-standing tropes about video games is that violent ones – like Call of Duty or Fortnite – can cause players to become more aggressive in the real world.

Neural approaches to Natural Language Generation (NLG) have been promising for goal-oriented dialogue. One of the challenges of productionizing these approaches, however, is the ability to control response quality, and ensure that generated responses are acceptable. We propose the use of a generate, filter, and rank framework, in which candidate responses are first filtered to eliminate unacceptable responses, and then ranked to select the best response. While acceptability includes grammatical correctness and semantic correctness, we focus only on grammaticality classification in this paper, and show that existing datasets for grammatical error correction don't correctly capture the distribution of errors that data-driven generators are likely to make. We release a grammatical classification and semantic correctness classification dataset for the weather domain that consists of responses generated by 3 data-driven NLG systems. We then explore two supervised learning approaches (CNNs and GBDTs) for classifying grammaticality. Our experiments show that grammaticality classification is very sensitive to the distribution of errors in the data, and that these distributions vary significantly with both the source of the response as well as the domain. We show that it's possible to achieve high precision with reasonable recall on our dataset.

Recommender system research suffers from a disconnect between the size of academic data sets and the scale of industrial production systems. In order to bridge that gap, we propose to generate large-scale user/item interaction data sets by expanding pre-existing public data sets. Our key contribution is a technique that expands user/item incidence matrices matrices to large numbers of rows (users), columns (items), and non-zero values (interactions). The proposed method adapts Kronecker Graph Theory to preserve key higher order statistical properties such as the fat-tailed distribution of user engagements, item popularity, and singular value spectra of user/item interaction matrices. Preserving such properties is key to building large realistic synthetic data sets which in turn can be employed reliably to benchmark recommender systems and the systems employed to train them. We further apply our stochastic expansion algorithm to the binarized MovieLens 20M data set, which comprises 20M interactions between 27K movies and 138K users. The resulting expanded data set has 1.2B ratings, 2.2M users, and 855K items, which can be scaled up or down.

Young isolated neutron stars (INS) most commonly manifest themselves as rotationally powered pulsars (RPPs) which involve conventional radio pulsars as well as gamma-ray pulsars (GRPs) and rotating radio transients (RRATs). Some other young INS families manifest themselves as anomalous X-ray pulsars (AXPs) and soft gamma-ray repeaters (SGRs) which are commonly accepted as magnetars, i.e.\ magnetically powered neutron stars with decaying super-strong fields. Yet some other young INS are identified as central compact objects (CCOs) and X-ray dim isolated neutron stars (XDINs) which are cooling objects powered by their thermal energy. Older pulsars, as a result of a previous long episode of accretion from a companion, manifest themselves as millisecond pulsars and more commonly appear in binary systems. We use Dirichlet process Gaussian mixture model (DPGMM), an unsupervised machine learning algorithm, for analyzing the distribution of these pulsar families in period $P$ and period derivative $\dot{P}$ parameter space. We compare the average values of the characteristic age, magnetic dipole field strength, surface temperature and proper motion of all discovered components. We verify that DPGMM is robust and provides hints for inferring relations between different classes of pulsars. We discuss the implications of our findings for the magnetothermal spin evolution models and fallback discs.

Noisy labels often occur in vision datasets, especially when they are issued from crowdsourcing or Web scraping. In this paper, we propose a new regularization method which enables one to learn robust classifiers in presence of noisy data. To achieve this goal, we augment the virtual adversarial loss with a Wasserstein distance. This distance allows us to take into account specific relations between classes by leveraging on the geometric properties of this optimal transport distance. Notably, we encode the class similarities in the ground cost that is used to compute the Wasserstein distance. As a consequence, we can promote smoothness between classes that are very dissimilar, while keeping the classification decision function sufficiently complex for similar classes. While designing this ground cost can be left as a problem-specific modeling task, we show in this paper that using the semantic relations between classes names already leads to good results.Our proposed Wasserstein Adversarial Training (WAT) outperforms state of the art on four datasets corrupted with noisy labels: three classical benchmarks and one real case in remote sensing image semantic segmentation.