Public privacy is at risk of mass scale invasion as an increasing use of Big Data and a surge in the amount of overt CCTV has left regulators struggling to keep pace, according to the UK's surveillance commissioner. Alongside the launch of a new three-year strategy on Tuesday, surveillance camera commissioner Tony Porter told the Guardian that the UK government is falling behind the pace at which new CCTV technology is being implemented. The 2017-2020 National Surveillance Camera Strategy for England and Wales, which was first proposed in October and has now passed through its consultation phase, aims to establish a code of practice for the use of CCTV equipment, including the use of body worn cameras and automatic number plate recognition. Under the strategy, disparate groups and regulators would be brought together in an attempt to create a set of coherent rules, limiting the use of surveillance cameras to times when it was deemed proportionate and necessary. However, Porter has expressed alarm that the increasing number of CCTV devices could lead to more invasive surveillance tactics than anticipated, as video footage is now being linked with data analysis tools like facial recognition or being cross referenced with other monitored personal data and activities.

From Google to Microsoft to Apple and possibly Samsung, AI is almost the buzzword in the smartphone space as well. But as these systems get more powerful and begin to do more, there is a need to make sure that they are not used by authoritarian regimes and then target certain populations. At the ongoing South by South West (SXSW) event in Texas, USA, Microsoft Research's Kate Crawford explained her point on view on the AI scene and how it makes for an ideal excuse to teach these humongous data systems the wrong ideas without any accountability. The Guardian reported the researcher's take on artificial intelligence, hinting at the rise of ultra-nationalism, right-wing authoritarianism and fascism, thanks to the widespread use of these systems. However, it is not the use of these systems, but the way in which human biases are encoded into them that leads to their misuse when they fall into the wrong hands.

An award-winning brewpub built with recycled materials as part of a "zero waste" mission. There is new life in the mountains of Tokushima Prefecture, in the neighboring towns of Kamiyama and Kamikatsu, even as depopulation afflicts most rural areas with rot. In Kamiyama, young people work remotely for tech companies or as artists in cooperative spaces. In Kamikatsu, the elderly test drones as part of their work harvesting leaves and flowers for use as garnishes in restaurants as far away as Europe. Prime Minister Shinzo Abe has stressed the need to revitalize rural areas as the country struggles with demographic decline.

Talespin is a young Delhi-based startup that wants to place chatbots inside stores. The idea, according to Talespin's product head Tanay Dixit, is to help bring some of the benefits of the online shopping experience - such as personalised recommendations - offline as well, at a low cost. The idea was actually born when Dixit tried to go shopping for a t-shirt at Shopper's Stop, and after being shown a lot of different ones, he finally found something he liked, only to learn it wasn't available in his size. He realised that if there was a system to browse through the store's entire catalogue like you would an e-commerce marketplace, then he would have been able to get a better idea more quickly, and it could even have allowed him to place an order for the out- of-stock item, to be delivered to him. And so that's exactly what he set out to build.

It's Monday morning, and you've just stepped into a driverless car with three colleagues to chair your first meeting of the day. You enter the office by placing your hand on a biometric scan, and, sensing your feelings as you start the working week, the lights and temperature are adjusted to improve your mood. Your virtual assistant orders you a coffee, which is brought to your desk by a co-bot working alongside you. This may sound like a sci-fi film, but these innovations could be coming to your workplace sooner than you think… Here, I explore the key technologies set to reshape the professional landscape by 2020, making the workplace of the future much more virtual, mobile, collaborative and flexible. Gartner predicts that 21 million new cars will be equipped with data connectivity this year alone.

Venture capital investments in artificial intelligence (AI), one of the biggest technological trends these days, are booming. The year 2016 saw almost ten times more funds invested in the space than 2012 did. Virtual and augmented reality (AR/VR), sometimes associated with the creative side of AI, also became a hot topic: More than $1.8 billion was invested in AR/VR, compared with just $86 million in 2012. Applying AI to solve tasks normally considered creative has historical precedent. For example, procedural generation has been used to draw textures, produce 3D models, and automatically generate large amounts of content in video games since the 1980s.

Chronic obstructive pulmonary disease (COPD) is a lung disease where early detection benefits the survival rate. COPD can be quantified by classifying patches of computed tomography images, and combining patch labels into an overall diagnosis for the image. As labeled patches are often not available, image labels are propagated to the patches, incorrectly labeling healthy patches in COPD patients as being affected by the disease. We approach quantification of COPD from lung images as a multiple instance learning (MIL) problem, which is more suitable for such weakly labeled data. We investigate various MIL assumptions in the context of COPD and show that although a concept region with COPD-related disease patterns is present, considering the whole distribution of lung tissue patches improves the performance. The best method is based on averaging instances and obtains an AUC of 0.742, which is higher than the previously reported best of 0.713 on the same dataset. Using the full training set further increases performance to 0.776, which is significantly higher (DeLong test) than previous results.

We present mlrMBO, a flexible and comprehensive R toolbox for model-based optimization (MBO), also known as Bayesian optimization, which addresses the problem of expensive black-box optimization by approximating the given objective function through a surrogate regression model. It is designed for both single- and multi-objective optimization with mixed continuous, categorical and conditional parameters. Additional features include multi-point batch proposal, parallelization, visualization, logging and error-handling. mlrMBO is implemented in a modular fashion, such that single components can be easily replaced or adapted by the user for specific use cases, e.g., any regression learner from the mlr toolbox for machine learning can be used, and infill criteria and infill optimizers are easily exchangeable. We empirically demonstrate that mlrMBO provides state-of-the-art performance by comparing it on different benchmark scenarios against a wide range of other optimizers, including DiceOptim, rBayesianOptimization, SPOT, SMAC, Spearmint, and Hyperopt.

In this paper, we aim at recovering an undirected weighted graph of $N$ vertices from the knowledge of a perturbed version of the eigenspaces of its adjacency matrix $W$. For instance, this situation arises for stationary signals on graphs or for Markov chains observed at random times. Our approach is based on minimizing a cost function given by the Frobenius norm of the commutator $\mathsf{A} \mathsf{B}-\mathsf{B} \mathsf{A}$ between symmetric matrices $\mathsf{A}$ and $\mathsf{B}$. In the Erd\H{o}s-R\'enyi model with no self-loops, we show that identifiability (i.e., the ability to reconstruct $W$ from the knowledge of its eigenspaces) follows a sharp phase transition on the expected number of edges with threshold function $N\log N/2$. Given an estimation of the eigenspaces based on a $n$-sample, we provide support selection procedures from theoretical and practical point of views. In particular, when deleting an edge from the active support, our study unveils that our test statistic is the order of $\mathcal O(1/n)$ when we overestimate the true support and lower bounded by a positive constant when the estimated support is smaller than the true support. This feature leads to a powerful practical support estimation procedure. Simulated and real life numerical experiments assert our new methodology.

Supervised learning has been very successful for automatic segmentation of images from a single scanner. However, several papers report deteriorated performances when using classifiers trained on images from one scanner to segment images from other scanners. We propose a transfer learning classifier that adapts to differences between training and test images. This method uses a weighted ensemble of classifiers trained on individual images. The weight of each classifier is determined by the similarity between its training image and the test image. We examine three unsupervised similarity measures, which can be used in scenarios where no labeled data from a newly introduced scanner or scanning protocol is available. The measures are based on a divergence, a bag distance, and on estimating the labels with a clustering procedure. These measures are asymmetric. We study whether the asymmetry can improve classification. Out of the three similarity measures, the bag similarity measure is the most robust across different studies and achieves excellent results on four brain tissue segmentation datasets and three white matter lesion segmentation datasets, acquired at different centers and with different scanners and scanning protocols. We show that the asymmetry can indeed be informative, and that computing the similarity from the test image to the training images is more appropriate than the opposite direction.