Billionaires in the world's tech capital Silicon Valley are reportedly preparing for the apocalypse by buying underground bunkers, guns, ammo and motorcycles. Fearful that artificial intelligence will displace so many jobs that there will be a revolt against those responsible for the technology, the are entrepreneurs readying themselves for doomsday like scenarios. Reid Hoffman, the co-founder of the professional social network, LinkedIn, told The New Yorker that he believes more than 50 per cent of billionaires in the Californian tech hub are preparing for the worst. "I own a couple of motorcycles. I have a bunch of guns and ammo.

Driving into Johannesburg, the skyline is often encased in a dome of smog. As Africa urbanises and industrialises, its economic growth is powered by coal, its vehicles by dirty fuels, and many still cook over paraffin and three-stone fires, burning their rubbish in the street. Researchers believe that more than 700,000 Africans die each year as a result of air pollution – and the number continues to grow. Few African cities even measure air pollution, but those that do routinely rank among the dirtiest on the planet. Nigeria is home to four of the worst, including Onitsha, which holds the dubious distinction of the Earth's most toxic air.

Scalable and effective exploration remains a key challenge in reinforcement learning (RL). While there are methods with optimality guarantees in the setting of discrete state and action spaces, these methods cannot be applied in high-dimensional deep RL scenarios. As such, most contemporary RL relies on simple heuristics such as epsilon-greedy exploration or adding Gaussian noise to the controls. This paper introduces Variational Information Maximizing Exploration (VIME), an exploration strategy based on maximization of information gain about the agent's belief of environment dynamics. We propose a practical implementation, using variational inference in Bayesian neural networks which efficiently handles continuous state and action spaces. VIME modifies the MDP reward function, and can be applied with several different underlying RL algorithms. We demonstrate that VIME achieves significantly better performance compared to heuristic exploration methods across a variety of continuous control tasks and algorithms, including tasks with very sparse rewards.

How to self-localize large teams of underwater nodes using only noisy range measurements? How to do it in a distributed way, and incorporating dynamics into the problem? How to reject outliers and produce trustworthy position estimates? The stringent acoustic communication channel and the accuracy needs of our geophysical survey application demand faster and more accurate localization methods. We approach dynamic localization as a MAP estimation problem where the prior encodes dynamics, and we devise a convex relaxation method that takes advantage of previous estimates at each measurement acquisition step; The algorithm converges at an optimal rate for first order methods. LocDyn is distributed: there is no fusion center responsible for processing acquired data and the same simple computations are performed for each node. LocDyn is accurate: experiments attest to a smaller positioning error than a comparable Kalman filter. LocDyn is robust: it rejects outlier noise, while the comparing methods succumb in terms of positioning error.

In coming years residential consumers will face real-time electricity tariffs with energy prices varying day to day, and effective energy saving will require automation - a recommender system, which learns consumer's preferences from her actions. A consumer chooses a scenario of home appliance use to balance her comfort level and the energy bill. We propose a Bayesian learning algorithm to estimate the comfort level function from the history of appliance use. In numeric experiments with datasets generated from a simulation model of a consumer interacting with small home appliances the algorithm outperforms popular regression analysis tools. Our approach can be extended to control an air heating and conditioning system, which is responsible for up to half of a household's energy bill.

Existing approaches to online convex optimization (OCO) make sequential one-slot-ahead decisions, which lead to (possibly adversarial) losses that drive subsequent decision iterates. Their performance is evaluated by the so-called regret that measures the difference of losses between the online solution and the best yet fixed overall solution in hindsight. The present paper deals with online convex optimization involving adversarial loss functions and adversarial constraints, where the constraints are revealed after making decisions, and can be tolerable to instantaneous violations but must be satisfied in the long term. Performance of an online algorithm in this setting is assessed by: i) the difference of its losses relative to the best dynamic solution with one-slot-ahead information of the loss function and the constraint (that is here termed dynamic regret); and, ii) the accumulated amount of constraint violations (that is here termed dynamic fit). In this context, a modified online saddle-point (MOSP) scheme is developed, and proved to simultaneously yield sub-linear dynamic regret and fit, provided that the accumulated variations of per-slot minimizers and constraints are sub-linearly growing with time. MOSP is also applied to the dynamic network resource allocation task, and it is compared with the well-known stochastic dual gradient method. Under various scenarios, numerical experiments demonstrate the performance gain of MOSP relative to the state-of-the-art.

As a postdoc, plant biologist Christopher Topp was not satisfied with the usual way of studying root development: growing plants on agar dishes and placing them on flatbed scanners to measure root lengths and angles. Five years later, the idea of using detailed imaging to study plant form and function has caught on. The use of drones and robots is also on the rise as researchers pursue the'quantified plant'--one in which each trait has been carefully and precisely measured from nearly every angle, from the length of its root hairs to the volatile chemicals it emits under duress. Such traits are known as an organism's phenotype, and researchers are looking for faster and more comprehensive ways of characterizing it. From February 10 to 14, scientists will gather in Tucson, Arizona, to compare their methods.

Time series (TS) occur in many scientific and commercial applications, ranging from earth surveillance to industry automation to the smart grids. An important type of TS analysis is classification, which can, for instance, improve energy load forecasting in smart grids by detecting the types of electronic devices based on their energy consumption profiles recorded by automatic sensors. Such sensor-driven applications are very often characterized by (a) very long TS and (b) very large TS datasets needing classification. However, current methods to time series classification (TSC) cannot cope with such data volumes at acceptable accuracy; they are either scalable but offer only inferior classification quality, or they achieve state-of-the-art classification quality but cannot scale to large data volumes. In this paper, we present WEASEL (Word ExtrAction for time SEries cLassification), a novel TSC method which is both scalable and accurate. Like other state-of-the-art TSC methods, WEASEL transforms time series into feature vectors, using a sliding-window approach, which are then analyzed through a machine learning classifier. The novelty of WEASEL lies in its specific method for deriving features, resulting in a much smaller yet much more discriminative feature set. On the popular UCR benchmark of 85 TS datasets, WEASEL is more accurate than the best current non-ensemble algorithms at orders-of-magnitude lower classification and training times, and it is almost as accurate as ensemble classifiers, whose computational complexity makes them inapplicable even for mid-size datasets. The outstanding robustness of WEASEL is also confirmed by experiments on two real smart grid datasets, where it out-of-the-box achieves almost the same accuracy as highly tuned, domain-specific methods.

As hard as we're trying, it's going to be a very long time before we're able to build a robotic insect that's anywhere near as capable or versatile as a real one. So for now, we rely on a cybernetics approach to get real insects to do our bidding instead. Over the past several years researchers have managed to steer large insects using electrical implants, a sort of brute-force method with limited real-world usefulness. R&D company Draper are hoping to overcome those limitations by creating a cybernetic dragonfly that combines "miniaturized navigation, synthetic biology, and neurotechnology." To steer the dragonflies, the Draper engineers are developing a way of genetically modifying the nervous system of the insects so they can respond to pulses of light.

The robot on an oil drillship in the Gulf of Mexico made it easier for Mark Rodgers to do his job stringing together heavy, dirty pipes. It could also be a reason he's not working there today. The Iron Roughneck, made by National Oilwell Varco Inc., automates the repetitive and dangerous task of connecting hundreds of segments of drill pipe as they're shoved through miles of ocean water and oil-bearing rock. The machine has also cut to two from three the need for roustabouts, estimates Rodgers, who took a job repairing appliances after being laid off from Transocean Ltd. "I'd love to go back offshore," he says. The odds are against him.