Hunt brilliantly describes how these processes work in "Sheep Logic"; a masterpiece. My chapter 20 contains a short version of the theory.

Lawmakers in the United States and Europe are demanding to know more about the company's privacy practices after a whistleblower said consultancy Cambridge Analytica improperly accessed data to target U.S. and British voters in elections.

If the draft of its AI strategy is anything to go by, the EU has yet to recognize the technology's epochal significance Saul Loeb/AFP via Getty Images Brussels is failing to grasp threats and opportunities of artificial intelligence. When the computer program AlphaGo beat the Chinese professional Go player Ke Jie in a three-part match, it didn't take long for Beijing to realize the implications. If algorithms can already surpass the abilities of a master Go player, it can't be long before they will be similarly supreme in the activity to which the classic board game has always been compared: war. As I've written before, the great conflict of our time is about who can control the next wave of technological development: the widespread application of artificial intelligence in the economic and military spheres. That's why it's so worrying that while China has been quick to react to the threats and opportunities of AI, the European Union -- if the draft of its AI strategy is anything to go by -- has yet to recognize the technology's epochal significance.

Chris Wylie, the former Cambridge Analytica employee turned whistleblower whose revelations about Facebook data being misused for political campaigning has wiped billions off the share price of the company in recent days and led to the FTC opening a fresh investigation, has suggested the scale of the data leak is substantially larger than has been reported so far. Giving evidence today, to a UK parliamentary select committee that's investigating the use of disinformation in political campaigning, Wylie said: "The 50 million number is what the media has felt safest to report -- because of the documentation that they can rely on -- but my recollection is that it was substantially higher than that. So my own view is it was much more than 50M." We've reached out to Facebook about Wylie's claim -- but at the time of writing the company had not provided a response. "There were several iterations of the Facebook harvesting project," Wylie also told the committee, fleshing out the process through which he says users' data was obtained by CA. "It first started as a very small pilot -- firstly to see, most simply, is this data matchable to an electoral register… We then scaled out slightly to make sure that [Cambridge University professor Alexsandr Kogan] could acquire data in the speed that he said he could [via a personality test app called thisisyourdigitallife deployed via Facebook's platform]. So the first real pilot of it was a sample of 10,000 people who joined the app -- that was in late May 2014. "That project went really well and that's when we signed a much larger contract with GSR [Kogan's company] in the first week of June… 2014.

Another weak spot was Nvidia, which fell 7.8 percent after the chipmaker temporarily suspended self-driving tests across the globe after an Uber Technologies Inc autonomous vehicle killed a woman. Investors rotated out of the tech sector, which had long outperformed the market on hopes of new technologies such as artificial intelligence (AI) and internet of things (IoT). "There is a sense that there will be more regulations on Facebook or FANG and that the cost of compliance will increase," said Nobuhiko Kuramochi, chief strategist at Mizuho Securities. The so-called FANG, a quartet of tech stocks that include Facebook, Amazon.com, Netflix and Alphabet, have been a darling of many investors.

Elias, the new language teacher at a Finnish primary school, has endless patience for repetition, never makes a pupil feel embarrassed for asking a question and can even do the'Gangnam Style' dance. Elias is also a robot. The language-teaching machine comprises a humanoid robot and mobile application, one of four robots in a pilot programme at primary schools in the southern city of Tampere. Pictured is Elias, a robot teaching children in a Finnish school. The robot is able to understand and speak 23 languages and is equipped with software that allows it to understand students' requirements and helps it to encourage learning.

Well-established methods for the solution of stochastic partial differential equations (SPDEs) typically struggle in problems with high-dimensional inputs/outputs. Such difficulties are only amplified in large-scale applications where even a few tens of full-order model runs are impracticable. While dimensionality reduction can alleviate some of these issues, it is not known which and how many features of the (high-dimensional) input are actually predictive of the (high-dimensional) output. In this paper, we advocate a Bayesian formulation that is capable of performing simultaneous dimension and model-order reduction. It consists of a component that encodes the high-dimensional input into a low-dimensional set of feature functions by employing sparsity-enforcing priors and a decoding component that makes use of the solution of a coarse-grained model in order to reconstruct that of the full-order model. Both components are represented with latent variables in a probabilistic graphical model and are simultaneously trained using Stochastic Variational Inference methods. The model is capable of quantifying the predictive uncertainty due to the information loss that unavoidably takes place in any model-order/dimension reduction as well as the uncertainty arising from finite-sized training datasets. We demonstrate its capabilities in the context of random media where fine-scale fluctuations can give rise to random inputs with tens of thousands of variables. With a few tens of full-order model simulations, the proposed model is capable of identifying salient physical features and produce sharp predictions under different boundary conditions of the full output which itself consists of thousands of components.

The recent successes of AI have captured the wildest imagination of both the scientific communities and the general public. Robotics and AI amplify human potentials, increase productivity and are moving from simple reasoning towards human-like cognitive abilities. Current AI technologies are used in a set area of applications, ranging from healthcare, manufacturing, transport, energy, to financial services, banking, advertising, management consulting and government agencies. The global AI market is around 260 billion USD in 2016 and it is estimated to exceed 3 trillion by 2024. To understand the impact of AI, it is important to draw lessons from it's past successes and failures and this white paper provides a comprehensive explanation of the evolution of AI, its current status and future directions.

Knowledge Transfer (KT) techniques tackle the problem of transferring the knowledge from a large and complex neural network into a smaller and faster one. However, existing KT methods are tailored towards classification tasks and they cannot be used efficiently for other representation learning tasks. In this paper a novel knowledge transfer technique, that is capable of training a student model that maintains the same amount of mutual information between the learned representation and a set of (possible unknown) labels as the teacher model, is proposed. Apart from outperforming existing KT techniques, the proposed method allows for overcoming several limitations of existing methods providing new insight into KT as well as novel KT applications, ranging from knowledge transfer from handcrafted feature extractors to {cross-modal} KT from the textual modality into the representation extracted from the visual modality of the data.

We introduce a Bayesian framework for inference with a supervised version of the Gaussian process latent variable model. The framework overcomes the high correlations between latent variables and hyperparameters by using an unbiased pseudo estimate for the marginal likelihood that approximately integrates over the latent variables. This is used to construct a Markov Chain to explore the posterior of the hyperparameters. We demonstrate the procedure on simulated and real examples, showing its ability to capture uncertainty and multimodality of the hyperparameters and improved uncertainty quantification in predictions when compared with variational inference.