Physicists commonly classify material phases as one or the other. Machine learning is a powerful tool for pattern recognition and thus could help identify phases of matter. However, machine learning needs a bridge to the quantum world, where the physics of atoms, electrons, and particles differs from that of larger objects or galaxies. Now, scientists have provided a bridge, which they call the quantum loop topography technique. This is a machine-learning algorithm based on neural networks.

The Justice Department will soon start trying to jam cellphones smuggled into federal prisons and used for criminal activity, part of a broader safety initiative that is also focused on preventing drones from airdropping contraband to inmates. Deputy Attorney General Rod J. Rosenstein told the American Correctional Association's conference in Orlando on Monday that, while the law prohibits cellphone use by federal inmates, the Bureau of Prisons confiscated 5,116 such phones in 2016, and preliminary numbers for 2017 indicate a 28 percent increase. "That is a major safety issue," he said in his speech. "Cellphones are used to run criminal enterprises, facilitate the commission of violent crimes and thwart law enforcement." When he was the U.S. attorney in Maryland, Rosenstein prosecuted an inmate who used a smuggled cellphone to order the murder of a witness.

Bitcoin and blockchain entered the consumer lexicon in a media frenzy that saw it hit $19,000 USD per coin (albeit for only 20 minutes) and led thousands of new hopefuls to Google'what is bitcoin and how will it make me rich?' And the WannaCry cyberattack saw consumers' cherished photos and documents held to ransom, crippling Britain's NHS and reminding everyone that cybersecurity isn't just for big corporations or spy agencies. With all of that in mind, along with Niels Bohr's observation that prediction is very difficult (especially about the future), here are our six predictions about how these key themes -- AI and Machine Learning (ML), IoT, security, blockchain -- are likely to begin to make a real difference in the way all of us interact with technology in 2018… Ever tried to use Siri offline? Suddenly, your knowledgeable little friend becomes rather dumb. In 2017, AI fed consumers content on Facebook or Netflix while assistants such as Siri, Alexa or Google Assistant identified what song was playing or what the weather would be like tomorrow.

President Vladimir V. Putin of Russia, in a surprise visit to that air base on Dec. 11, declared that combat operations were winding down and that the Russian military would stage a "significant withdrawal." It was at least the second time he had made such an announcement since March 2016. Mr. Putin faces a presidential election this March, and although he is expected to win easily, polls indicate that Russians are increasingly disgruntled about the country's military presence in Syria. Please verify you're not a robot by clicking the box. You must select a newsletter to subscribe to.

HTC is upgrading its headsets for exploring virtual worlds. HTC says the new Vive Pro has better resolution and audio and weighs less than its existing VR model. The Taiwanese company hasn't yet revealed cost or shipping dates. The Vive competes with Facebook's Oculus among high-end systems. They require powerful personal computers to run and haven't been as widely used as cheaper headsets that use smartphones, including Samsung's Gear VR and Google's Daydream.

This article describes Team Kernel Glitches' solution to the National Institute of Justice's (NIJ) Real-Time Crime Forecasting Challenge. The goal of the NIJ Real-Time Crime Forecasting Competition was to maximize two different crime hotspot scoring metrics for calls-for-service to the Portland Police Bureau (PPB) in Portland, Oregon during the period from March 1, 2017 to May 31, 2017. Our solution to the challenge is a spatiotemporal forecasting model combining scalable randomized Reproducing Kernel Hilbert Space (RKHS) methods for approximating Gaussian processes with autoregressive smoothing kernels in a regularized supervised learning framework. Our model can be understood as an approximation to the popular log-Gaussian Cox Process model: we discretize the spatiotemporal point pattern and learn a log intensity function using the Poisson likelihood and highly efficient gradient-based optimization methods. Model hyperparameters including quality of RKHS approximation, spatial and temporal kernel lengthscales, number of autoregressive lags, bandwidths for smoothing kernels, as well as cell shape, size, and rotation, were learned using crossvalidation. Resulting predictions exceeded baseline KDE estimates by 0.157. Performance improvement over baseline predictions were particularly large for sparse crimes over short forecasting horizons.

Recent studies show that widely used deep neural networks (DNNs) are vulnerable to carefully crafted adversarial examples. Many advanced algorithms have been proposed to generate adversarial examples by leveraging the $\mathcal{L}_p$ distance for penalizing perturbations. Researchers have explored different defense methods to defend against such adversarial attacks. While the effectiveness of $\mathcal{L}_p$ distance as a metric of perceptual quality remains an active research area, in this paper we will instead focus on a different type of perturbation, namely spatial transformation, as opposed to manipulating the pixel values directly as in prior works. Perturbations generated through spatial transformation could result in large $\mathcal{L}_p$ distance measures, but our extensive experiments show that such spatially transformed adversarial examples are perceptually realistic and more difficult to defend against with existing defense systems. This potentially provides a new direction in adversarial example generation and the design of corresponding defenses. We visualize the spatial transformation based perturbation for different examples and show that our technique can produce realistic adversarial examples with smooth image deformation. Finally, we visualize the attention of deep networks with different types of adversarial examples to better understand how these examples are interpreted.

A research team from a hospital in Oxford have created artificial intelligence that can diagnose scans for lung cancer and heart disease. The AI system will reportedly help in saving billions of dollars by helping to diagnose the diseases much earlier. NHS hospitals can avail the technology for free, beginning summer 2018, and AI could help in saving the NHS. "There is about £2.2bn spent on pathology services in the NHS. You may be able to reduce that by 50%," said immunologist Sir John Bell.

In December, NASA announced two finalist concepts for a robotic mission that will launch in the mid-2020s. The first is the Comet Astrobiology Exploration Sample Return (CAESAR), which would send a fairly conventional spacecraft over to a comet to grab a chunk of its nucleus and bring it back to Earth. That's cool and all, but we're much more excited about the second finalist concept: Dragonfly, from the Johns Hopkins University Applied Physics Lab (APL), a quad octocopter that would explore Saturn's moon Titan from the air. The idea is that it would work like a planetary rover, except that it would fly instead of drive, allowing it to cover much more ground at the risk of, you know, crashing. We've seen lots of drones that can do amazing things, and also lots of drones that crash very, very badly while trying to do amazing things. Sending a fully autonomous flying robot to an alien world over a billion kilometers away and expecting it to fly around for a couple years without any human intervention seems extraordinarily ambitious, so we checked in with APL to see exactly what they're working on.

Automotive supplier Bosch wants to help guide drivers to vacant parking spots in more than a dozen U.S. cities this year. The German company says it's been testing its "community-based parking" initiative in Stuttgart and other German cities and will launch it later this year in as many as 20 U.S. cities, including Los Angeles, Miami and Boston. The company says it will be working with automakers on the initiative but didn't say which ones. As cars drive by, they will automatically recognize and measure gaps between parked cars and transmit that data to a digital map. The company has been pushing a number of smart-city projects, including internet-connected sensors to monitor pollution, allergens and flooding.