Earthquakes occurred here in 1857, 1881, 1901, 1922, 1934, and 1966, suggesting a pattern in which quakes occur every 22 years give or take a few years. Geologists know that as a quake approaches, the gouge material begins to fail, emitting groans and cracks as it shears--a kind of seismic chatter. "We show that by listening to the acoustic signal emitted by a laboratory fault, machine learning can predict the time remaining before it fails with great accuracy," they say. The first and most obvious question it raises is whether the same technique could predict real earthquakes accurately.

Microsoft wrapped up last week, Apple is coming soon, and Google is, well, today! We'll be reporting live from the I/O keynote, which starts this afternoon. We also take a look at the origins of the sex robot. This year's Google developer event kicks off today. We're expecting to learn more about the future of Android, Chrome and other projects.

Innovations in digitization, analytics, artificial intelligence, and automation are creating performance and productivity opportunities for business and the economy, even as they reshape employment and the future of work. Rapid technological advances in digitization and data and analytics have been reshaping the business landscape, supercharging performance, and enabling the emergence of new business innovations and new forms of competition. At the same time, the technology itself continues to evolve, bringing new waves of advances in robotics, analytics, and artificial intelligence (AI), and especially machine learning. Together they amount to a step change in technical capabilities that could have profound implications for business, for the economy, and more broadly, for society. Some companies are gaining a competitive edge with their use of data and analytics, which can enable faster and larger-scale evidence-based decision making, insight generation, and process optimization.

A former Google China president and now venture capitalist says Elon Musk uses shiny cars and the promise of medical implants as bait for his real goals: Distributing energy away from traditional power companies and turning humans into cyborgs. First, says Kai-Fu Lee, were Tesla cars. "By selling us fancy, beautiful Teslas--luxury cars that none of us can say no to, it seems to have changed to distributed energy," Lee, the CEO of Sinovation Ventures, told Quartz in an interview today. As Tesla CEO, Musk has acquired the solar energy startup SolarCity that he previously helped lead as chairman, then he began sharing a vision where a battery in the home stores energy from the sun (preferably using SolarCity's new solar panels). That energy will be used to power the home and charge electric cars.

Multi-armed bandits are a quintessential machine learning problem requiring the balancing of exploration and exploitation. While there has been progress in developing algorithms with strong theoretical guarantees, there has been less focus on practical near-optimal finite-time performance. In this paper, we propose an algorithm for Bayesian multi-armed bandits that utilizes value-function-driven online planning techniques. Building on previous work on UCB and Gittins index, we introduce linearly-separable value functions that take both the expected return and the benefit of exploration into consideration to perform n-step lookahead. The algorithm enjoys a sub-linear performance guarantee and we present simulation results that confirm its strength in problems with structured priors. The simplicity and generality of our approach makes it a strong candidate for analyzing more complex multi-armed bandit problems.

As the world begins to turn away from fossil fuels and depend increasingly on renewable resources, the energy sector is presented with a problem. Renewables are simply not as reliable as oil and gas, as they are largely dependent on weather conditions such as sunny skies and windy days. In a world where we become fully dependent on renewables, there is concern that supply may not always be able to meet demand. This supply problem is compounded with the complications of individuals, businesses, and municipalities becoming small-scale energy producers themselves by way of solar panels and individual storage units connected to the grid. These producer-consumers, having varying and unpredictable patterns of individual production and consumption create instability on shared grids.

Gut reaction: You won't want to hide the HTC U11 in a protective case. HTC's new flagship phone is just too stunning to conceal behind a chintzy polycarb shell. Imbued with HTC's new "liquid" design aesthetic, the U11 has an impossibly glossy finish, evoking the T-1000 Terminator for those old enough to remember that robot assassin. From a basic specs perspective, the U11 is a solid but not remarkable Android 7.1 phone. It includes a state-of-the-silicon-art Snapdragon 835 processor running up to 2.45 GHz, a 3,000 mAh battery, and a 5.5-inch quad HD (2560x1440) display.

Machine learning has emerged as an invaluable tool in many research areas. In the present work, we harness this power to predict highly accurate molecular infrared spectra with unprecedented computational efficiency. To account for vibrational anharmonic and dynamical effects -- typically neglected by conventional quantum chemistry approaches -- we base our machine learning strategy on ab initio molecular dynamics simulations. While these simulations are usually extremely time consuming even for small molecules, we overcome these limitations by leveraging the power of a variety of machine learning techniques, not only accelerating simulations by several orders of magnitude, but also greatly extending the size of systems that can be treated. To this end, we develop a molecular dipole moment model based on environment dependent neural network charges and combine it with the neural network potentials of Behler and Parrinello. Contrary to the prevalent big data philosophy, we are able to obtain very accurate machine learning models for the prediction of infrared spectra based on only a few hundreds of electronic structure reference points. This is made possible through the introduction of a fully automated sampling scheme and the use of molecular forces during neural network potential training. We demonstrate the power of our machine learning approach by applying it to model the infrared spectra of a methanol molecule, n-alkanes containing up to 200 atoms and the protonated alanine tripeptide, which at the same time represents the first application of machine learning techniques to simulate the dynamics of a peptide. In all these case studies we find excellent agreement between the infrared spectra predicted via machine learning models and the respective theoretical and experimental spectra.

Several problems requiere the combination of temporal and spatial reasoning under uncertainty, such as wind prediction for electricity generation in wind farms. In this work we propose a probabilistic spatial-temporal model (PSTM) focused on prediction problems, based on two common properties of these scenarios: sparsity and multivariable mutual information. The proposed spatial-temporal model is essentially a Bayesian network that represents the dependencies between a target variable of interest and a subset of predictor variables in different times and spaces. We developed an algorithm for learning the structure of the model based on a stochastic search of the optimal subset of predictor variables. The proposed model has been applied for wind prediction at different locations in Mexico, using information from several locations at different times. The PSTM is evaluated in terms of predictive accuracy for different time horizons — 1 to 24 hours; and compared to a dynamic Bayesian network (DBN) developed for wind prediction. The performance of the PSTM is in general competitive, and in most cases superior to the DBN.

Bayesian optimization through Gaussian process regression is an effective method of optimizing an unknown function for which every measurement is expensive. It approximates the objective function and then recommends a new measurement point to try out. This recommendation is usually selected by optimizing a given acquisition function. After a sufficient number of measurements, a recommendation about the maximum is made. However, a key realization is that the maximum of a Gaussian process is not a deterministic point, but a random variable with a distribution of its own. This distribution cannot be calculated analytically. Our main contribution is an algorithm, inspired by sequential Monte Carlo samplers, that approximates this maximum distribution. Subsequently, by taking samples from this distribution, we enable Thompson sampling to be applied to (armed-bandit) optimization problems with a continuous input space. All this is done without requiring the optimization of a nonlinear acquisition function. Experiments have shown that the resulting optimization method has a competitive performance at keeping the cumulative regret limited.