If 2016 seemed politically tumultuous, 2017 promises to be equally tumultuous on the technology front. The pace of change is accelerating at a dizzying rate, with profound implications for the way we work, play and communicate. So what are the big technology trends to watch out for in 2017? Cybersecurity will undoubtedly be the dominant theme of 2017, as all tech innovations could be undermined by data thefts, fraud and cyber propaganda. Forget Kim Kardashian, it's hacking that could break the internet - and much more besides.

Human-robot relationships are a running theme in pop culture, from the cylons of Battlestar Galactica to Spike Jonze's film Her and last year's hit show Westworld. But that kind of scenario might not be science fiction much longer. Romance between humans and machines is already nearing the realm of the possible. This year, the California company Abyss Creations plans to start selling a new generation of high-tech sex robots -- dolls that can actually speak and respond to touch. And according to artificial intelligence expert Dr. David Levy, in a few generations, we won't just be having sex with robots, we'll be marrying them.

We propose a family of probabilistic description logics (DLs) that are derived in a principled way from Halpern's probabilistic first-order logic. The resulting probabilistic DLs have a two-dimensional semantics similar to temporal DLs and are well-suited for representing subjective probabilities. We carry out a detailed study of reasoning in the new family of logics, concentrating on probabilistic extensions of the DLs ALC and EL, and showing that the complexity ranges from PTime via ExpTime and 2ExpTime to undecidable.

A major challenge in the training of recurrent neural networks is the so-called vanishing or exploding gradient problem. The use of a norm-preserving transition operator can address this issue, but parametrization is challenging. In this work we focus on unitary operators and describe a parametrization using the Lie algebra $\mathfrak{u}(n)$ associated with the Lie group $U(n)$ of $n \times n$ unitary matrices. The exponential map provides a correspondence between these spaces, and allows us to define a unitary matrix using $n^2$ real coefficients relative to a basis of the Lie algebra. The parametrization is closed under additive updates of these coefficients, and thus provides a simple space in which to do gradient descent. We demonstrate the effectiveness of this parametrization on the problem of learning arbitrary unitary operators, comparing to several baselines and outperforming a recently-proposed lower-dimensional parametrization. We additionally use our parametrization to generalize a recently-proposed unitary recurrent neural network to arbitrary unitary matrices, using it to solve standard long-memory tasks.

Picture a self-driving car test in your head and you probably see an engineer or two scrutinizing data... and no one else. Everyday people, if they're present at all, tend to be relegated to the back seat. Volvo is trying something different: it just revealed that it's conducting autonomous vehicle tests with an ordinary family, the Hains from Gothenburg, Sweden. The four-person household is convenient for marketing, of course (we care about people!), but they serve an important purpose: they'll help Volvo understand how non-engineers deal with self-driving tech. How do they react when the car switches between manual and autonomous modes, and what do they do at those times when they aren't taking the wheel?

It is well known that the global optimum of a MDP with finite state and action sets can be obtained through methods based on dynamic programming. Unfortunately, these techniques are known to suffer from the curse of dimensionality, which makes them infeasible for many real-world problems of interest. As a result, most research in the reinforcement learning and control theory literature has focused on obtaining approximate or locally optimal solutions. There exists a broad spectrum of such techniques, including approximate dynamic programming methods, tree search methods, local trajectory-optimization techniques, such as differential dynamic programming and iLQG, and policy search methods. In this talk I shall provide an introduction to policy search methods, which are a family of algorithms that have proven extremely popular in recent years, and which have numerous desirable properties that make them attractive in practice.

When did electricity take over from steam in the UK? When did football replace cricket as the most popular sport? And what year did women start to become more frequently mentioned in the press? Specifically, a new paper by a team artificial intelligence researchers at the University of Bristol that used AIto analyse the news from 100 different British regional newspapers over the past 150 years. The team of academics, led by professor Nello Cristianini, collaborated closely with the company findmypast, which is digitising historical newspapers from the British Library as part of their British Newspaper Archive project.

And, it was first published on Dataconomy on Dec 21, 2016; I'm reposting here for ease of access.] In recent years, AI (artificial intelligence) researchers have finally cracked problems that they've worked on for decades, from Go to human-level speech recognition. A key piece was the ability to gather and learn on mountains of data, which pulled error rates past the success line. In short, big data has transformed AI, to an almost unreasonable level. Blockchain technology could transform AI too, in its own particular ways. Some applications of blockchains to AI are mundane, like audit trails on AI models. Some appear almost unreasonable, like AI that can own itself -- AI DAOs. All of them are opportunities. This article will explore these applications. Before we discuss applications, let's first review what's different about blockchains compared to traditional big-data distributed databases like MongoDB. We can think of blockchains as "blue ocean" databases: they escape the "bloody red ocean" of sharks competing in an existing market, opting instead to be in a blue ocean of uncontested market space.

Patient Number Two was born to first-time parents, late 20s, white. The pregnancy was normal and the birth uncomplicated. But after a few months, it became clear something was wrong. The child had ear infection after ear infection and trouble breathing at night. He was small for his age, and by his fifth birthday, still hadn't spoken.

In a Harvard Business School classroom in Boston, MA, robots are on the rise. MBA students are trying to crack a case study on the self-driving cars pioneered by Tesla, Google, and Uber. What is the potential for robots to reshape our roads? And what are the challenges and opportunities of entering that business? This is a case that David Yoffie, professor of international business administration, believes is essential reading for tomorrow's business leaders.