Although science and the public celebrated the amazing predictive power of the new machine learningmethods, many researchers are left with some unease, simply because prediction does not imply causation. The ability to uncover causal relations is, however, at the core of most questions concerning the effects of particular policies, medical treatments, marketing campaigns, businessdecisions, etc. (see Athey, 2017, for a recent discussion). The recently rapidly expanding causal machine learning literature holds great promise for the improved estimation of causal effects by merging the statistics and econometrics literature oncausality with the supervised statistical and machine learning (ML) literature focussing on prediction. The classical causality literature clarifies the conditions needed for being able to estimate causal effects. It also shows how to transform a counterfactual causal problem into specific prediction problems (e.g., Imbens and Wooldridge, 2009). The latter literature on ML provides tools that can be highly effective in solving prediction problems (e.g.

This paper develops a novel graph convolutional network (GCN) framework for fault location in power distribution networks. The proposed approach integrates multiple measurements at different buses while takes system topology into account. The effectiveness of the GCN model is corroborated by the IEEE 123-bus benchmark system. Simulation results show that the GCN model significantly outperforms other widely-used machine learning schemes with very high fault location accuracy. In addition, the proposed approach is robust to measurement noise and errors, missing entries, as well as multiple connection possibilities. Finally, data visualization results of two competing neural networks are presented to explore the mechanism of GCN's superior performance.

We consider the problem of training speech recognition systems without using any labeled data, under the assumption that the learner can only access to the input utterances and a phoneme language model estimated from a non-overlapping corpus. We propose a fully unsupervised learning algorithm that alternates between solving two sub-problems: (i) learn a phoneme classifier for a given set of phoneme segmentation boundaries, and (ii) refining the phoneme boundaries based on a given classifier. To solve the first sub-problem, we introduce a novel unsupervised cost function named Segmental Empirical Output Distribution Matching, which generalizes the work in (Liu et al., 2017) to segmental structures. For the second sub-problem, we develop an approximate MAP approach to refining the boundaries obtained from Wang et al. (2017). Experimental results on TIMIT dataset demonstrate the success of this fully unsupervised phoneme recognition system, which achieves a phone error rate (PER) of 41.6%. Although it is still far away from the state-of-the-art supervised systems, we show that with oracle boundaries and matching language model, the PER could be improved to 32.5%.This performance approaches the supervised system of the same model architecture, demonstrating the great potential of the proposed method.

Researchers at China's Zhejiang University published a study last year that showed many of the most popular smart speakers and smartphones, equipped with digital assistants, could be easily tricked into being controlled by hackers. They used a technique called DolphinAttack, which translates voice commands into ultrasonic frequencies that are too high for the human ear to recognize. While the commands may go unheard by humans, the low-frequency audio commands can be picked up, recovered and then interpreted by speech recognition systems. The team were able to launch attacks, which are higher than 20kHz, by using less than £2.20 ($3) of equipment which was attached to a Galaxy S6 Edge. They used an external battery, an amplifier, and an ultrasonic transducer.

The Army used a cutting-edge Israeli anti-drone system to defeat the unmanned aerial vehicle (UAV) that brought misery to hundreds of thousands of people at Gatwick airport. The British Army bought six'Drone Dome' systems for £15.8 million in 2018 and the technology is used in Syria to destroy ISIS UAVs. Police had been seen on Thursday with an off-the-shelf DJI system that tracks drones made by that manufacturer and shows officers where the operator is (DJI is the most popular commercial drone brand.) However, the drone used at Gatwick is thought to have been either hacked or an advanced non-DJI drone, which rendered the commercial technology used by the police useless. At that point, the Army's'Drone Dome' system made by Rafael was called in.

With two months to go until Samsung unveils its next flagship phone – presumably called the Galaxy S10 – we already know almost everything there is to know. A slew of leaks and rumours mean little has been left to the imagination about the iPhone rival, with the latest leak revealing the lengths the South Korean electronic giant has gone to avoid the "notch" design. For several years, smartphone manufacturers have been getting closer and closer to making an all-screen device, though necessary front-facing technologies like cameras and sensors have proved a major obstacle to achieving this goal. Apple's answer was to include a notch at the top of the screen, which it unveiled pm the iPhone X in 2017 to widespread acclaim. Not everyone was impressed and Samsung took the opportunity to mock its chief rival.

Slack has banned accounts with alleged links to countries under US sanctions, despite some users of the workplace chat app claiming they only visited them on holiday. In a message to affected users, Slack said accounts would be closed "effective immediately," in order to comply with laws and regulations imposed by the US government. Sanctioned countries and regions include Cuba, Iran, North Korea, Syria and the Crimea region of Ukraine. "In order to comply with export control and economic sanctions laws and regulations promulgated by the US Department of Commerce and the US Department of Treasury, Slack prohibits unauthorised use of its products and services in certain sanctioned countries," the message to affected users stated. "We've identified your team/ account as originating from one of those countries and are closing the account effective immediately."

In 2017, Google began a project known as PAIR, People AI Research with the goal of of building an AI that truly treats everyone equally. This project, however socially and technologically forward thinking it may be, we see a very different and more sinister approach to AI of the future on the other side of the world. Today, China has an estimated two hundred million surveillance cameras, about four times as many as in the US. An indispensable tool for the Chinese police, AI powered cameras and facial recognition smart-glasses scan citizens' faces, gathering data along the way. In the face of AI, many of us may be fearing for our jobs, but many more live in fear of their very lives.

Computer scientists at Stanford University created the first AI, an algorithm called Forage. Using a dataset of 60,000 recipes, they trained Forage to craft new recipes using whatever you tell it is in your fridge. When given a list of typical holiday leftovers to work with, Forage came up with recipes for a Spanish potato casserole, turkey croquettes, and a spicy seafood casserole. The BBC reports that they were, respectively, "well-received," "not too bad to eat," and "about as close to a bout of food poisoning as we were prepared to get." After reading the recipe for that last one, in fact, they couldn't even bring themselves to cook it.

Last July, photographer Reuben Wu and a crew of around 30 people hiked from the Peruvian city of Huaraz, nestled in the Cordillera Blanca region of the Andes, to the 16,000-foot-high Pastoruri glacier. The hike took around four hours and the crew arrived after sunset, finding the melting glacier lit only by a full moon. These Stitched Photos of Greenland's Icebergs Are Sew Great Wu has shot conceptual landscape photography in some of the world's most remote locations--East Java, Patagonia, Chile's Atacama Desert, Norway's Svalbard Archipelago--but this shoot, part of a mini-documentary about Wu's photography done as part of a Coors Light ad campaign, gave him the opportunity to highlight global warming by photographing a fast-receding glacier, one of the last in South America. "There were parts of the glacier where you could see evidence of pretty extreme breakdown and melting of the snow," Wu says. "Parts of the glacier no longer had the epic, jagged chunks of ice."