Earlier promises of progress turned out to be premature. The green light could be delayed again if proponents can't overcome nagging security concerns on the part of local or national law-enforcement agencies. Proposed projects also may end up stymied if Federal Aviation Administration managers don't find creative ways around legislative and regulatory restrictions such as those mandating pilot training for manned aircraft. But some proponents of delivery and other drone applications "think they might be ready to operate this summer," Jay Merkle, a senior FAA air-traffic control official, said during a break at an unmanned-aircraft conference in Baltimore last week that highlighted the agency's pro-business approach. At least 10 FAA-approved pilot programs for various drone initiatives--some likely including package delivery--are slated to start by May.

Retrieving the most similar objects in a large-scale database for a given query is a fundamental building block in many application domains, ranging from web searches, visual, cross media, and document retrievals. State-of-the-art approaches have mainly focused on capturing the underlying geometry of the data manifolds. Graph-based approaches, in particular, define various diffusion processes on weighted data graphs. Despite success, these approaches rely on fixed-weight graphs, making ranking sensitive to the input affinity matrix. In this study, we propose a new ranking algorithm that simultaneously learns the data affinity matrix and the ranking scores. The proposed optimization formulation assigns adaptive neighbors to each point in the data based on the local connectivity, and the smoothness constraint assigns similar ranking scores to similar data points. We develop a novel and efficient algorithm to solve the optimization problem. Evaluations using synthetic and real datasets suggest that the proposed algorithm can outperform the existing methods.

Deep neural networks (DNNs) enable innovative applications of machine learning like image recognition, machine translation, or malware detection. However, deep learning is often criticized for its lack of robustness in adversarial settings (e.g., vulnerability to adversarial inputs) and general inability to rationalize its predictions. In this work, we exploit the structure of deep learning to enable new learning-based inference and decision strategies that achieve desirable properties such as robustness and interpretability. We take a first step in this direction and introduce the Deep k-Nearest Neighbors (DkNN). This hybrid classifier combines the k-nearest neighbors algorithm with representations of the data learned by each layer of the DNN: a test input is compared to its neighboring training points according to the distance that separates them in the representations. We show the labels of these neighboring points afford confidence estimates for inputs outside the model's training manifold, including on malicious inputs like adversarial examples--and therein provides protections against inputs that are outside the models understanding. This is because the nearest neighbors can be used to estimate the nonconformity of, i.e., the lack of support for, a prediction in the training data. The neighbors also constitute human-interpretable explanations of predictions. We evaluate the DkNN algorithm on several datasets, and show the confidence estimates accurately identify inputs outside the model, and that the explanations provided by nearest neighbors are intuitive and useful in understanding model failures.

Early stopping of iterative algorithms is a widely-used form of regularization in statistics, commonly used in conjunction with boosting and related gradient-type algorithms. Although consistency results have been established in some settings, such estimators are less well-understood than their analogues based on penalized regularization. In this paper, for a relatively broad class of loss functions and boosting algorithms (including L2-boost, LogitBoost and AdaBoost, among others), we exhibit a direct connection between the performance of a stopped iterate and the localized Gaussian complexity of the associated function class. This connection allows us to show that local fixed point analysis of Gaussian or Rademacher complexities, now standard in the analysis of penalized estimators, can be used to derive optimal stopping rules. We derive such stopping rules in detail for various kernel classes, and illustrate the correspondence of our theory with practice for Sobolev kernel classes.

Japan is answering the problem of too many wild boars creatively: by scaring them off with robotic wolves. The wolves, which have fangs, fur and red eyes, are being implemented to help farmers protect their crops from the growing wild boar population, Earther reports. The wolves were tested last summer, and the product will now be produced and available all over Japan. The wild boar population in Japan's northern regions has been increasing. Sometimes called the'Super Monster Wolf', the invention runs on batteries that are solar-rechargeable, and its range is greater than half a mile.

A federal judge has tossed the Massachusetts attorney general's lawsuit against President Donald Trump's administration over rules allowing more employers to opt out of providing no-cost birth control to women.

Blaming violence on video games over-simplifies a deeply complicated issue. If you're wondering whether the video game your child is playing is appropriate, there's a long-standing rating system in place to guide you. But that system, established in 1994 after Senate hearings into violence in video games, is "good but not perfect," says Jeff Haynes, senior editor for video games at Common Sense Media, a non-profit group that curates its own library of ratings and reviews of games, movies apps, TV shows and other content. What that means is parents need to be especially vigilant when it comes to assessing whether the games their children play are appropriate. "The biggest thing we constantly try to push is know your kids and know the content your kids are playing and get involved with what your kids are interested in," Haynes says.

BOSTON (Reuters) - A federal judge on Monday rejected a lawsuit by Massachusetts' attorney general challenging new rules by President Donald Trump's administration that make it easier for employers to avoid providing insurance that covers women's birth control.

Last month the Congressional Subcommittee on Information Technology began a three-part series of hearings to break through the myths and the hype to gain a real understanding of Artificial Intelligence (AI) and the role it can play in the Federal government. While the first hearing focused on industry and academic experts, Wednesday's hearing saw testimony exclusively from government leaders, including representatives from Defense Advanced Research Projects Agency (DARPA), General Services Administration (GSA), National Science Foundation (NSF), and Department of Homeland Security (DHS). During the hearing one point was raised that bears repeating–AI isn't science fiction, its science reality. "When people hear'artificial intelligence,' their minds often wander to the realm of science fiction," said Keith Nakasone, deputy assistant commissioner, acquisition operations, Office of Information Technology Category, GSA. "There is also a belief that AI is in the future, rather than in the present. Concerted effort by policymakers of all levels to help change this narrative will be critical in promoting acceptance and adoption of AI by more and more entities."

Don't be surprised if you see a drone outside on your doorstep this summer. Federal regulators want to begin using drones for'limited package deliveries' as soon as within the next few months, according to the Wall Street Journal. Officials have been working with Silicon Valley tech giants and aerospace companies to develop proposals, rewrite regulations and address safety concerns, as part of an effort to make the technology a reality. A drone delivers an Amazon package to customers in Germany. The Federal Aviation Administration (FAA) made similar promises last year, but their efforts were stymied by growing concerns from local and national law-enforcement agencies.