The World Economic Forum has put together a list of the top 10 emerging technologies that will change our lives. The list includes nanosensors that will circulate through the human body, a battery that will be able to power an entire town and socially aware artificial intelligence that will track our finances and health. These are not far-flung visions, according to the forum. They are technologies that are on the cusp of having a meaningful impact. "Horizon scanning for emerging technologies is crucial to staying abreast of developments that can radically transform our world, enabling timely expert analysis in preparation for these disruptors," said Bernard Meyerson, chairman of the World Economic Forum council that compiled the list of the top 10 emerging technologies in 2016.

Linear dimensionality reduction techniques are powerful tools for image analysis as they allow the identification of important features in a data set. In particular, nonnegative matrix factorization (NMF) has become very popular as it is able to extract sparse, localized and easily interpretable features by imposing an additive combination of nonnegative basis elements. Nonnegative matrix underapproximation (NMU) is a closely related technique that has the advantage to identify features sequentially. In this paper, we propose a variant of NMU that is particularly well suited for image analysis as it incorporates the spatial information, that is, it takes into account the fact that neighboring pixels are more likely to be contained in the same features, and favors the extraction of localized features by looking for sparse basis elements. We show that our new approach competes favorably with comparable state-of-the-art techniques on synthetic, facial and hyperspectral image data sets.

I just noticed that Justin Domke has a blog -- He's one of the strongest researchers in the field of graphical models. I first came across his dissertation when looking for a way to improve loopy-Belief Propagation based training. His thesis gives one such idea -- instead of maximizing the fit of an intractable model, and using BP as intermediate step, maximize the fit of BP marginals directly. This makes sense since approximate (BP-based) marginals are what you ultimately use. If you run BP for k steps, then likelihood of the BP-approximated model is tractable to minimize -- calculation of gradient is very similar to k steps of loopy BP.

The infrastructure and tools necessary for large-scale data analytics, formerly the exclusive purview of experts, are increasingly available. Whereas a knowledgeable data-miner or domain expert can rightly be expected to exercise caution when required (for example, around fallacious conclusions supposedly supported by the data), the nonexpert may benefit from some judicious assistance. This article describes an end-to-end learning framework that allows a novice to create models from data easily by helping structure the model building process and capturing extended aspects of domain knowledge. By treating the whole modeling process interactively and exploiting high-level knowledge in the form of an ontology, the framework is able to aid the user in a number of ways, including in helping to avoid pitfalls such as data dredging. Prudence must be exercised to avoid these hazards as certain conclusions may only be supported if, for example, there is extra knowledge which gives reason to trust a narrower set of hypotheses. This article adopts the solution of using higher-level knowledge to allow this sort of domain knowledge to be used automatically, selecting relevant input attributes, and thence constraining the hypothesis space. We describe how the framework automatically exploits structured knowledge in an ontology to identify relevant concepts, and how a data extraction component can make use of online data sources to find measurements of those concepts so that their relevance can be evaluated. To validate our approach, models of four different problem domains were built using our implementation of the framework. Prediction error on unseen examples of these models show that our framework, making use of the ontology, helps to improve model generalization.

PRO: Chris Bishop, director of Microsoft Research in Cambridge, said earlier this year that he believes the hyperbole around the AI risks could jeopardise any future developments that may in fact assist humanity. "Any scenario in which AI is an existential threat to humanity is not just around the corner," he told the Guardian. Referring to the views of high-profile cynics like professor Stephen Hawking, Bishop said: "I think they must be talking decades away for those comments to make any sense. Right now we are in control of that technology and we can make lots of choices about the paths that we follow." Oren Etzioni, chief executive of the Allen Institute for AI and professor of computer science at the University of Washington, meanwhile says the popular dystopian vision of AI is wrong because it "equates intelligence with autonomy".

When I read the headlines Friday about the fatal crash of a Tesla vehicle while in self-driving mode, I immediately thought about Three Mile Island. It's not that Tesla's autopilot mode is the vehicular equivalent of a nuclear meltdown. As the company would very much like you to note, self-driving cars are doing better, statistically speaking, than human drivers. Tesla says autopilot was used for 130 million miles worth of driving before this fatal crash. Human-driven cars in the U.S. have 1.08 fatal crashes for every 100 million miles.

This paper considers global optimization with a black-box unknown objective function that can be non-convex and non-differentiable. Such a difficult optimization problem arises in many real-world applications, such as parameter tuning in machine learning, engineering design problem, and planning with a complex physics simulator. This paper proposes a new global optimization algorithm, called Locally Oriented Global Optimization (LOGO), to aim for both fast convergence in practice and finite-time error bound in theory. The advantage and usage of the new algorithm are illustrated via theoretical analysis and an experiment conducted with 11 benchmark test functions. Further, we modify the LOGO algorithm to specifically solve a planning problem via policy search with continuous state/action space and long time horizon while maintaining its finite-time error bound.

The Model S tells drivers to keep their hands on the wheel; if they release their grip it alerts drivers and eventually slows down. Of course, a driver who's distracted for even an instant can cause deadly harm on the road. "The system also makes frequent checks to ensure that the driver's hands remain on the wheel and provides visual and audible alerts if hands-on is not detected," Tesla wrote in a blog post announcing the federal probe. "It then gradually slows down the car until hands-on is detected again. We do this to ensure that every time the feature is used, it is used as safely as possible."

Could Kevin Bacon have saved us from the 2008 financial crisis? But the network science behind six degrees of Kevin Bacon just well may have. According to the famous saying, every movie actor is separated from Kevin Bacon by six degrees of separation or less, going from co-star to co-star (actually most are separated from Bacon by only three degrees). Actors form a "small-world" network, meaning it takes a surprisingly small number of connections to get from any one member to any other. Natural and man-made small-world networks of all kinds are extremely common: The electric power grid of the western United States, the neural network of the nematode worm C. elegans, the Internet, protein and gene networks in biology, citations in scientific papers, and most social networks are small. Most of these small networks use hubs, or nodes with an especially large number of links to other nodes.