Your son is hanging out with friends behind the VR theatre when someone pulls a knife: they're being robbed. He hits a panic button on his phone, and a nearby drone is instantly dispatched. Less than 30 seconds later it's in the alley, sirens wailing and lights flashing, recording everything to be transmitted to the police. The technology for that scenario exists today. Add a dash of image recognition, or take it another step and put a taser (or a miniaturized version of the heat ray) on the device, and you have the automation of emergency response services.

With flashy AI technology like IBM's Watson and Google's driverless cars stealing headlines and outperforming their human competitors, it's clear that our economy is bracing for a fundamental shift in how we perform work. What's less obvious, however, is exactly what the workplace of the future will look like. A pair of Oxford researchers recently estimated that 47 percent of the total U.S. employment is at risk of being eliminated. On the other end of the spectrum, Mercedes announced it is trading out some of its production robots for human labor -- the machines could not keep up with the increasing options for customization. While these two camps continue to argue, in this article we'll explore three robotic trends that the prevailing media have missed in their coverage of the future of jobs -- trends that will hold true if we continue this automation trajectory.

Minecraft has become a worldwide phenomenon in recent years, and its blocky universe be used to hone the next generation of artificial intelligence? Computer scientists at Microsoft Research think so, and have been using the game's universe to train an AI'agent' to learn how to do things, such as climb a mountain, using the same types of resources a human has when they learn a new task. Much to Stephen Hawking's chagrin, AI has come on leaps and bounds in recent years, and computers can now understand speech and translate it, as well as being able recognise images and write captions about them. But computers still aren't very good at what researchers call general intelligence, which is more similar to the nuanced and complex way humans learn and make decisions. This is where AIX, a platform developed by Katja Hofmann and her colleagues in Microsoft's Cambridge lab, comes in. The system is a mod for the Java version of Minecraft and code that helps artificial intelligence agents sense and act within the game environment.

If you have always wanted a cat, but can't choose between a plump and lazy tabby or a playful and slinky Siamese, a new online service could help. Experts have come up with an algorithm to match cats looking for a home with would-be doting owners, based on their personality. PawsLikeMe has been likened to eHarmony and hopes to make people's choices'more than fur deep' so less cats will need re-homing again. The Orlando Florida-based initiative built an algorithm to match dogs with new owners last year and since then, two million people in the US have used the'first ever human-to-pet matching algorithm' to find a furry friend. It claims to be 90 per cent accurate in predicting people-to-pet compatibility. Now it's raising money on Indiegogo to collect enough funds to roll out a version for cats.

In 1981, science essayist Jeremy Bernstein wrote a piece for The New Yorker that touched upon a historic backgammon game two years earlier, in which reigning champ Luigi Villa lost to a computer. It was the first time an artificial-intelligence program had defeated a world champion at a board or card game. In the essay, Bernstein wrote: "What does this mean for us, for our sense of uniqueness and worth -- especially as machines evolve whose output we can less and less distinguish from our own?" He might have asked that decades ago, but the question is now more relevant than ever. Google's AlphaGo recently won four out of five matches against Go master Lee Se-dol.

This article describes methods for machine learning using bootstrap samples and parallel processing to model very large volumes of data in short periods of time. The R programming language includes many packages for machine learning different types of data. Three of these packages include Support Vector Machines (SVM) [1], Generalized Linear Models (GLM) [2], and Adaptive Boosting (AdaBoost) [3]. While all three packages can be highly accurate for various types of classification problems, each package performs very differently when modeling (i.e. In particular, model fitting for Generalized Linear Models execute in much shorter periods of time than either Support Vector Machines or Adaptive Boosting.

On Friday afternoon, video game news site Kotaku dropped a fascinating story. Sony is rumoured to be working on an upgraded version of PlayStation 4 complete with support for the emerging 4K resolution, and more processing power to cope with the demands of virtual reality. Kotaku cited unnamed developers as the source of its report, and claims to have overheard discussions between programmers about the new "PlayStation 4.5" during the Game Developers Conference in San Francisco last week (where the hardware was allegedly being revealed in closed demos to key studios). It's not clear whether the new format is an upgrade that attaches to the existing PS4 or a completely new replacement, but Sony has, of course, said that it won't comment on rumours and speculation. So let's say it's happening.

Every time we post about autonomous delivery drones, we have to point out that despite the promises implied by overproduced and optimistic videos, the drones are simply not capable of autonomous navigation in complex environments. Same goes for those camera drones that promise to follow you: the videos inevitably show them following skiers on wide open slopes, surfers on the wide open sea, or other people doing things very far away from inconvenient obstacles like trees. So far, we've only seen a tiny handful of drones capable of dynamically detecting and avoiding obstacles at a useful speed. Qualcomm and UPenn have been working on some, and MIT has that speedy tree-avoiding fixed-wing drone. A Silicon Valley company called Skydio, founded by a team of researchers from MIT and Google X's Project Wing, have posted a video that shows a drone following people jogging and biking while autonomously avoiding tree trunks and branches.

Imagine you're the designer for an advertising campaign for a furniture store. That campaign will run on desktops, and in email newsletters, but it will also need to live on tablets and phones. You'll need different photos for different devices, and suddenly, creating one campaign is more like creating four. As screens (and screen sizes) proliferate, this is an increasingly common problem. At Adobe's digital marketing conference in Las Vegas, one of many new features the creative tools company announced is particularly poised to offer relief to anyone working in branding or marketing.

It is the goal of the proposed workshop to bring together researchers from the fields of machine learning and statistical physics in order to discuss the new challenges originating from dynamical data. Such data are modeled using a variety of approaches such as dynamic belief networks, continuous time analogues of these – as often used for disordered spin systems in statistical physics –, coupled stochastic differential equations for continuous random variables etc. The workshop provides a forum for exploring possible synergies between the inference and learning approaches developed for the various models. The experience from joint advances in the equilibrium domain suggests that there is much unexplored scope for progress on dynamical data.