Jobs have become a core Brexit issue. Would leaving the EU be good or bad for job creation? Why are migrants taking so many of the new jobs being created in the UK, and what are the implications for the freedom of movement for labour? But the changing nature of work is an issue that runs throughout the developed world, far beyond the UK and Europe. We see this through the prism of our relationship with Europe, for the UK and Germany have become the two strongest job markets in this time zone and have accordingly been sucking in labour from elsewhere.

CAIRO – A French ship searching the Mediterranean has detected black box signals from a missing EgyptAir flight in the waters between the Greek island of Crete and the Egyptian coast, a development that could help solve the mystery of why the aircraft crashed into the sea last month, killing all 66 on board. The discovery, announced Wednesday, could help guide search teams to the wreckage and the flight's data and cockpit voice recorders, which if retrieved unharmed could reveal whether a mechanical fault or a hijacking or bomb caused the disaster. In the two weeks since Flight 804 disappeared from radar en route to Cairo from Paris, only small pieces of debris and human remains have been retrieved from the crash site. No terrorist group has claimed responsibility, though Egypt's civil aviation minister, Sherif Fathi, has said terrorism is a more likely cause than equipment failure or some other catastrophic event. The flight recorders will be critical to determining whether the disaster was caused by an accident or a deliberate act.

"If we can do it in the right and thoughtful way, we are always open to it," said Pichai at the Code conference here. "I care about serving consumers everywhere." Google pulled out of mainland China and moved its Chinese-language search engine to Hong Kong in 2010 after a series of cyber attacks on Google originated in the country. Google also said it would stop censoring search results in China. The controversial move cut Google off from the fast-growing Chinese market, one that's been courted by rival Facebook and constitutes the second-biggest market for Apple.

IBM's Watson, an Artificial Intelligence (AI) system designed to answer questions posed in natural language. Few people know as much about travel search technology or the bookings process as WayBlazer's Terrell Jones. From his start as a travel agent to his role as CIO of SABRE and his involvement in the successful launch of Travelocity and Kayak, Jones has played a key role in shaping travel search. We met with him during SITA's Air Transport IT Summit (ATIS) in Barcelona last month to better understand how this technology is evolving. According to Jones, there's an opportunity for the airline industry to make stronger connections with its customers by applying intelligent technology.

Facebook has long relied on human eyes to monitor the 350 million photos that are uploaded daily. Now, the social media giant is using artificial intelligence to shoulder some of the weight and says the technology has flagged more offensive photos than its human users. Photos deemed offensive include content that is hateful, pornographic or violent and the AI scans every image to determine if they violate Facebook guidelines prior to releasing them on the site. Facebookis using artificial intelligence to shoulder some of the weight and says it has flagged more offensive photos than its human users. Some 350 million photos are upload to Facebook every day.

Bill Gates has proclaimed the'AI dream is finally arriving' - despite admitting it could be a major concern for the future of humanity. 'The dream is finally arriving,' Gates said, speaking with wife Melinda Gates on Wednesday at the Code Conference in Southern California. 'The dream is finally arriving,' Gates said, speaking at the Code Conference in Southern California. 'This is what it was all leading up to.' Gates said enough progress has been made to ensure that in the next 10 years there will be robots to do tasks like driving and warehouse work as well as machines that can outpace humans in certain areas of knowledge, according to recode. He also suggested a pair of books that people should read, including Nick Bostrom's book on superintelligence and Pedro Domingos' 'The Master Algorithm.'

Boston Dynamics' robots are at their best on shaky ground. The wobbly BigDog, its larger LS3 sibling, and the humanoid Atlas all shamble forward and recover from falls. The company, for sale by its owner Google, is now looking for a buyer, and now it appears they may have finally found a firm place to land: Toyota. The Tokyo-based company has nursing care and medical robots in development, in addition to working on self-driving cars. Toyota could add up to 300 personnel to its robotics division with the acquisitions, the report said.

LASSICAL unsupervised learning such as Principal Components Analysis (PCA) and Independent Component Analysis (ICA) is useful to design artificial sensory systems and to understand the organization of natural sensory systems. On the artificial side, examples include representations/transforms for image coding [6]-[9] and image categorization [10], [11]. On the natural side, examples include the analysis of visual cortex [12]-[16]. PCA and ICA obtain basis of the space according to different optimization criteria. These basis functions can be interpreted as linear sensors: the projection of data onto these basis represents the response of the set of sensors. PCA defines a sensor hierarchy: for example, an image sensory system made out of principal directions with highest eigenvalues minimizes the image reconstruction error [6], [7]. In ICA, the basis is intended to provide responses as independent as possible, which is equivalent to design a sensory system that maximizes the transmitted information (infomax) [17], [18].

We present a novel k-way high-dimensional graphical model called the Generalized Root Model (GRM) that explicitly models dependencies between variable sets of size k > 2---where k = 2 is the standard pairwise graphical model. This model is based on taking the k-th root of the original sufficient statistics of any univariate exponential family with positive sufficient statistics, including the Poisson and exponential distributions. As in the recent work with square root graphical (SQR) models [Inouye et al. 2016]---which was restricted to pairwise dependencies---we give the conditions of the parameters that are needed for normalization using the radial conditionals similar to the pairwise case [Inouye et al. 2016]. In particular, we show that the Poisson GRM has no restrictions on the parameters and the exponential GRM only has a restriction akin to negative definiteness. We develop a simple but general learning algorithm based on L1-regularized node-wise regressions. We also present a general way of numerically approximating the log partition function and associated derivatives of the GRM univariate node conditionals---in contrast to [Inouye et al. 2016], which only provided algorithm for estimating the exponential SQR. To illustrate GRM, we model word counts with a Poisson GRM and show the associated k-sized variable sets. We finish by discussing methods for reducing the parameter space in various situations.

Kernel methods are one of the mainstays of machine learning, but the problem of kernel learning remains challenging, with only a few heuristics and very little theory. This is of particular importance in methods based on estimation of kernel mean embeddings of probability measures. For characteristic kernels, which include most commonly used ones, the kernel mean embedding uniquely determines its probability measure, so it can be used to design a powerful statistical testing framework, which includes nonparametric two-sample and independence tests. In practice, however, the performance of these tests can be very sensitive to the choice of kernel and its lengthscale parameters. To address this central issue, we propose a new probabilistic model for kernel mean embeddings, the Bayesian Kernel Embedding model, combining a Gaussian process prior over the Reproducing Kernel Hilbert Space containing the mean embedding with a conjugate likelihood function, thus yielding a closed form posterior over the mean embedding. The posterior mean of our model is closely related to recently proposed shrinkage estimators for kernel mean embeddings, while the posterior uncertainty is a new, interesting feature with various possible applications. Critically for the purposes of kernel learning, our model gives a simple, closed form marginal pseudolikelihood of the observed data given the kernel hyperparameters. This marginal pseudolikelihood can either be optimized to inform the hyperparameter choice or fully Bayesian inference can be used.