If you've had a window seat next to the wing of an airplane, you've probably watched as flaps on the wing engage and disengage as a plane takes off and lands. That's because in each phase of flight -- take off, landing, cruising and maneuvering -- the ideal wing parameters vary. Until now, we've made do by modifying rigid wings with hinged surfaces. But imagine if the entire wing could change shape -- that's what researchers led by NASA and MIT are working towards. In a paper in the journal Smart Materials and Structures, the research team explains how it has radically redesigned the airplane wing.

Money laundering accounts for up to 5% of global GDP - or $2tn (£1.5tn) - every year, says the United Nations Office on Drugs and Crime. So banks and law enforcement agencies are turning to artificial intelligence (AI) to help combat the growing problem. Money laundering, so-called after gangster Al Capone's practice of hiding criminal proceeds in cash-only laundromats in the 1920s, is a huge and growing problem. "Dirty" money is "cleaned" by passing it through layers of seemingly legitimate banks and businesses and using it to buy properties, businesses, expensive cars, works of art - anything that can be sold on for new cash. And one of the ways criminals do this is called "smurfing".

The countdown has begun to the launch of one of the world's boldest attempts to tackle air pollution. From next Monday, thousands of drivers face paying a new charge to enter central London. The aim is to deter the dirtiest vehicles in an effort to reduce diseases and premature deaths. The initiative comes as scientists say the impacts of air pollution are more serious than previously thought. The mayor of London, Sadiq Khan, told the BBC that the threat of air pollution, which is mostly invisible to the naked eye, is "a public health emergency".

We use static object data to improve success detection for stacking objects on and nesting objects in one another. Such actions are necessary for certain robotics tasks, e.g., clearing a dining table or packing a warehouse bin. However, using an RGB-D camera to detect success can be insufficient: same-colored objects can be difficult to differentiate, and reflective silverware cause noisy depth camera perception. We show that adding static data about the objects themselves improves the performance of an end-to-end pipeline for classifying action outcomes. Images of the objects, and language expressions describing them, encode prior geometry, shape, and size information that refine classification accuracy. We collect over 13 hours of egocentric manipulation data for training a model to reason about whether a robot successfully placed unseen objects in or on one another. The model achieves up to a 57% absolute gain over the task baseline on pairs of previously unseen objects.

Understanding of the effect of a particular treatment or a policy pertains to many areas of interest -- ranging from political economics, marketing to health-care and personalized treatment studies. In this paper, we develop a non-parametric, model-free test for detecting the effects of treatment over time that extends widely used Synthetic Control tests. The test is built on counterfactual predictions arising from many learning algorithms. In the Neyman-Rubin potential outcome framework with possible carry-over effects, we show that the proposed test is asymptotically consistent for stationary, beta mixing processes. We do not assume that class of learners captures the correct model necessarily. We also discuss estimates of the average treatment effect, and we provide regret bounds on the predictive performance. To the best of our knowledge, this is the first set of results that allow for example any Random Forest to be useful for provably valid statistical inference in the Synthetic Control setting. In experiments, we show that our Synthetic Learner is substantially more powerful than classical methods based on Synthetic Control or Difference-in-Differences, especially in the presence of non-linear outcome models.

Missing data are a concern in many real world data sets and imputation methods are often needed to estimate the values of missing data, but data sets with excessive missingness and high dimensionality challenge most approaches to imputation. Here we show that appropriate feature selection can be an effective preprocessing step for imputation, allowing for more accurate imputation and subsequent model predictions. The key feature of this preprocessing is that it incorporates uncertainty: by accounting for uncertainty due to missingness when selecting features we can reduce the degree of missingness while also limiting the number of uninformative features being used to make predictive models. We introduce a method to perform uncertainty-aware feature selection (UAFS), provide a theoretical motivation, and test UAFS on both real and synthetic problems, demonstrating that across a variety of data sets and levels of missingness we can improve the accuracy of imputations. Improved imputation due to UAFS also results in improved prediction accuracy when performing supervised learning using these imputed data sets. Our UAFS method is general and can be fruitfully coupled with a variety of imputation methods.

Broadband customers who are having internet problems are about to start getting refunds – without even having to ask. At the moment, only about one in seven people who have internet or landline problems such as repairs, installations or missed engineer appointments are given any kind of compensation from the companies responsible, according to regulator Ofcom. Even if they do, the amounts are usually small. But now customers will find themselves being given those refunds automatically, for any kind of broadband problems, Ofcom said. We'll tell you what's true.

If you're inclined to puns, you might say medical samples are the lifeblood of hospital systems. But if you actually work with them, you know they're more of a headache. Because the same road traffic that keeps you from getting home keeps the couriers charged with moving these tissue and blood samples, collected by the millions daily and often in urgent need of analysis, from completing their missions. So it makes a lot of sense that when the FAA decided to sanction the first revenue-generating drone delivery scheme in the US, it went with one that promises to speed up that process, run by UPS and autonomous drone technology firm Matternet. It makes sense from the tech perspective, too: The cargo is extremely lightweight and compact, allowing the companies involved to focus on the delivery processes and mechanisms rather than trying to manage unwieldy payloads.

It's probably fair to say that in the history of politicking, few politicians have publicly declared what to do about America's crumbling malls, or how to provide free marriage counseling for all, or how to make filing taxes fun. But Andrew Yang, who's gunning to be the Democratic presidential candidate in 2020, certainly has--and those are the more minor concerns among a dizzying list of 80 policy positions on his campaign website. It's an indication that Yang is running a rather methodical, data-driven, science-happy campaign. He's applying that approach to more standard issue problems, like labor, climate change, and the economy, but giving them a decidedly tech-forward approach: how (and why) we should define robots, what use might geoengineering have in saving the planet, and should the government embrace universal basic income and give every American a $1,000 check. Yang talked with WIRED about all this and more in a recent interview.

Process (MDP) using Gaussian processes. In their work, they assumed the transition model is known and that there exists I. INTRODUCTION a predefined safety function. Both of these assumptions can Guaranteeing safety is a vital issue for many modern be quite restrictive when the system is going to operate in robotics systems, such as unmanned aerial vehicles (UAVs), unknown environments. In our work, we plan to address autonomous cars, or domestic robots [1], [2], [3]. One both of these challenges by considering unknown transition approach is to attempt to specify all potential scenarios models, and no access to a predefined safety function.