LONDON – Warning: the Kremlin is trying to split the West by spreading "altered facts," conducting blackmail and setting up front organizations, the U.S. State Department said -- in 1981. So-called active measures were common during the Cold War, when the U.S. and the Soviet Union sought to unify and divide Europe with equal urgency. Now those tactics appear to be back, retooled for the digital age as President Vladimir Putin embraces the even older Russian foreign policy tradition of "derzhavnost," or "great powerness." Fears of Russian interference are rampant across the continent. Already reeling from Brexit, the European Union faces a string of key elections starting next month in the Netherlands, then in France and Germany.

Last year, it was reported that supercomputer IBM Watson diagnosed a rare form of leukaemia in a patient at a University of Tokyo-affiliated hospital whose case had baffled her medical team. The cloud-based, artificial intelligence-powered supercomputer is capable of cross-referencing and analysing data from tens of millions of oncology papers from research institutes all over the world. From vast volumes of data, it can instantly pull out the information it needs, much faster than humans can. The University of Tokyo reported that the 60-year-old Japanese woman was correctly diagnosed in just 10 minutes by Watson, after her genetic data was cross-referenced with the computer's own database. More and more, health technologies originally viewed as futuristic – like virtual avatars and chatbots – have become reality.

From January 2005 through February 2015, shares of Apple Inc. closed at a record 264 times. But that run could end this week. Lee Jae-yong will be questioned based on findings uncovered in the past few weeks, the special prosecutor's office said. Verizon Communications will start selling unlimited data plans on Monday, the first time it has offered such a service since 2011 and a sign that intense competition is forcing the nation's largest carrier to respond. In a crowded and broadly successful weekend box office, "The Lego Batman Movie" beat out the sequels "Fifty Shades Darker" and "John Wick: Chapter 2." .

What will you watch in a driverless car? (Picture: Shutterstock) For most of automotive history, time spent in cars has been predominately spent driving, talking and listening to the radio. But by 2021, driverless cars will be in mass production and racing towards the mainstream, releasing those in the so-called "driving" seat of much of their previous preoccupations. Research by Harvard Health Watch found the average American spends 101 minutes driving each day. So, what are they going to do to avoid getting bored on dull, long journeys? This is unlikely to be a problem for long, with entertainment technology experts already dreaming up masterplans for what drivers will do when they no longer drive.

In an expanded edition published in 1988, they responded to claims that their 1969 conclusions significantly reduced funding for neural network research: "Our version is that progress had already come to a virtual halt because of the lack of adequate basic theories… by the mid-1960s there had been a great many experiments with perceptrons, but no one had been able to explain why they were able to recognize certain kinds of patterns and not others."

The inference of correlated signal fields with unknown correlation structures is of high scientific and technological relevance, but poses significant conceptual and numerical challenges. To address these, we develop the correlated signal inference (CSI) algorithm within information field theory (IFT) and discuss its numerical implementation. To this end, we introduce the free energy exploration (FrEE) strategy for numerical information field theory (NIFTy) applications. The FrEE strategy is to let the mathematical structure of the inference problem determine the dynamics of the numerical solver. FrEE uses the Gibbs free energy formalism for all involved unknown fields and correlation structures without marginalization of nuisance quantities. It thereby avoids the complexity marginalization often impose to IFT equations. FrEE simultaneously solves for the mean and the uncertainties of signal, nuisance, and auxiliary fields, while exploiting any analytically calculable quantity. Finally, FrEE uses a problem specific and self-tuning exploration strategy to swiftly identify the optimal field estimates as well as their uncertainty maps. For all estimated fields, properly weighted posterior samples drawn from their exact, fully non-Gaussian distributions can be generated. Here, we develop the FrEE strategies for the CSI of a normal, a log-normal, and a Poisson log-normal IFT signal inference problem and demonstrate their performances via their NIFTy implementations.

We exhibit a strong link between frequentist PAC-Bayesian risk bounds and the Bayesian marginal likelihood. That is, for the negative log-likelihood loss function, we show that the minimization of PAC-Bayesian generalization risk bounds maximizes the Bayesian marginal likelihood. This provides an alternative explanation to the Bayesian Occam's razor criteria, under the assumption that the data is generated by an i.i.d distribution. Moreover, as the negative log-likelihood is an unbounded loss function, we motivate and propose a PAC-Bayesian theorem tailored for the sub-gamma loss family, and we show that our approach is sound on classical Bayesian linear regression tasks.

We study the performance of data-driven, a priori and random approaches to label space partitioning for multi-label classification with a Gaussian Naive Bayes classifier. Experiments were performed on 12 benchmark data sets and evaluated on 5 established measures of classification quality: micro and macro averaged F1 score, Subset Accuracy and Hamming loss. Data-driven methods are significantly better than an average run of the random baseline. In case of F1 scores and Subset Accuracy - data driven approaches were more likely to perform better than random approaches than otherwise in the worst case. There always exists a method that performs better than a priori methods in the worst case. The advantage of data-driven methods against a priori methods with a weak classifier is lesser than when tree classifiers are used.

This paper tackles the reduction of redundant repeating generation that is often observed in RNN-based encoder-decoder models. Our basic idea is to jointly estimate the upper-bound frequency of each target vocabulary in the encoder and control the output words based on the estimation in the decoder. Our method shows significant improvement over a strong RNN-based encoder-decoder baseline and achieved its best results on an abstractive summarization benchmark.

Several deep learning models have been proposed for question answering. However, due to their single-pass nature, they have no way to recover from local maxima corresponding to incorrect answers. To address this problem, we introduce the Dynamic Coattention Network (DCN) for question answering. The DCN first fuses co-dependent representations of the question and the document in order to focus on relevant parts of both. Then a dynamic pointing decoder iterates over potential answer spans. This iterative procedure enables the model to recover from initial local maxima corresponding to incorrect answers. On the Stanford question answering dataset, a single DCN model improves the previous state of the art from 71.0% F1 to 75.9%, while a DCN ensemble obtains 80.4% F1.