In many security and healthcare systems, the detection and diagnosis systems use a sequence of sensors/tests. Each test outputs a prediction of the latent state and carries an inherent cost. However, the correctness of the predictions cannot be evaluated since the ground truth annotations may not be available. Our objective is to learn strategies for selecting a test that gives the best tradeoff between accuracy and costs in such unsupervised sensor selection (USS) problems. Clearly, learning is feasible only if ground truth can be inferred (explicitly or implicitly) from the problem structure. It is observed that this happens if the problem satisfies the'Weak Dominance' (WD) property [1]. We set up the USS problem as a stochastic partial monitoring problem and develop an algorithm with sub-linear regret under the WD property. We argue that our algorithm is optimal and evaluate its performance on problem instances generated from synthetic and real-world datasets.

Bayesian optimal design of experiments (BODE) has been successful in acquiring information about a quantity of interest (QoI) which depends on a black-box function. BODE is characterized by sequentially querying the function at specific designs selected by an infill-sampling criterion. However, most current BODE methods operate in specific contexts like optimization, or learning a universal representation of the black-box function. The objective of this paper is to design a BODE for estimating the statistical expectation of a physical response surface. This QoI is omnipresent in uncertainty propagation and design under uncertainty problems. Our hypothesis is that an optimal BODE should be maximizing the expected information gain in the QoI. We represent the information gain from a hypothetical experiment as the Kullback-Liebler (KL) divergence between the prior and the posterior probability distributions of the QoI. The prior distribution of the QoI is conditioned on the observed data and the posterior distribution of the QoI is conditioned on the observed data and a hypothetical experiment. The main contribution of this paper is the derivation of a semi-analytic mathematical formula for the expected information gain about the statistical expectation of a physical response. The developed BODE is validated on synthetic functions with varying number of input-dimensions. We demonstrate the performance of the methodology on a steel wire manufacturing problem.

Latent factor models are increasingly popular for modeling multi-relational knowledge graphs. By their vectorial nature, it is not only hard to interpret why this class of models works so well, but also to understand where they fail and how they might be improved. We conduct an experimental survey of state-of-the-art models, not towards a purely comparative end, but as a means to get insight about their inductive abilities. To assess the strengths and weaknesses of each model, we create simple tasks that exhibit first, atomic properties of binary relations, and then, common inter-relational inference through synthetic genealogies. Based on these experimental results, we propose new research directions to improve on existing models.

The prevalence of networked sensors and actuators in many real-world systems such as smart buildings, factories, power plants, and data centers generate substantial amounts of multivariate time series data for these systems. The rich sensor data can be continuously monitored for intrusion events through anomaly detection. However, conventional threshold-based anomaly detection methods are inadequate due to the dynamic complexities of these systems, while supervised machine learning methods are unable to exploit the large amounts of data due to the lack of labeled data. On the other hand, current unsupervised machine learning approaches have not fully exploited the spatial-temporal correlation and other dependencies amongst the multiple variables (sensors/actuators) in the system for detecting anomalies. In this work, we propose an unsupervised multivariate anomaly detection method based on Generative Adversarial Networks (GANs). Instead of treating each data stream independently, our proposed MAD-GAN framework considers the entire variable set concurrently to capture the latent interactions amongst the variables. We also fully exploit both the generator and discriminator produced by the GAN, using a novel anomaly score called DR-score to detect anomalies by discrimination and reconstruction. We have tested our proposed MAD-GAN using two recent datasets collected from real-world CPS: the Secure Water Treatment (SWaT) and the Water Distribution (WADI) datasets. Our experimental results showed that the proposed MAD-GAN is effective in reporting anomalies caused by various cyber-intrusions compared in these complex real-world systems.

We present a probabilistic deep learning methodology that enables the construction of predictive data-driven surrogates for stochastic systems. Leveraging recent advances in variational inference with implicit distributions, we put forth a statistical inference framework that enables the end-to-end training of surrogate models on paired input-output observations that may be stochastic in nature, originate from different information sources of variable fidelity, or be corrupted by complex noise processes. The resulting surrogates can accommodate high-dimensional inputs and outputs and are able to return predictions with quantified uncertainty. The effectiveness our approach is demonstrated through a series of canonical studies, including the regression of noisy data, multi-fidelity modeling of stochastic processes, and uncertainty propagation in high-dimensional dynamical systems.

While there have been a few massive surveillance startups in China that have raised funds on the back of computer vision advances, there's seemed to be less fervor outside of that market. Tel Aviv-based AnyVision is aiming to leverage its computer vision chops in tracking people and objects to create some pretty clear utility for the enterprise world. After announcing a $27 million Series A in mid-2018, the computer vision startup is bringing Lightspeed Venture Partners into the raise, closing out the round at $43 million. "When you have a company with the technology AnyVision has, and the market need that I'm hearing from across industries, what you need to do is push the gas pedal and build an organization which can monetize and take on this opportunity to grow massively," Lightspeed partner Raviraj Jain told TechCrunch. Right now the 200-person company has its eyes on the security and identity markets as it aims to bring its computer vision technology into more industry-tailored solutions.

Xiaomi is a phone brand you may have never heard of, but you will. Although it was possible to buy its handsets for some time if you looked hard enough, its first official move into the UK came last November with a highly attractive, innovative phone called the Mi 8 Pro. This week, its second flagship phone goes on sale, the Mi Mix 3, and it's even more eye-catching. No, not like those tiddly Samsung feature phones from the nineties, this phone slides only about 1cm, enough to reveal the camera hidden behind. Which, in turn, means the phone's front really is all screen – well, technically it has a 93.4 per cent screen-to-body ratio.

ActiveProtective developed a wearable that protects the hips of older adults using wearable airbags, alerting caregivers, monitors behaviors, and promotes safer mobility. BitLumens provides decentralized power to rural communities--for example, offering farmers solar home systems which they can pay in installments--which are not connected to the power line using the blockchain to provide them with a credit score. Civic Eagle built cloud platform using artificial intelligence (AI) to reduce the time and cost of identifying, tracking, and analyzing important legislation and regulations for organizations that need to respond to quickly changing policy environments. Einride vows to disrupt the unsustainable transportation industry, making the movement of goods more intelligent (emission-free, safe, and cost-efficient) by building interconnected, all-electric, autonomous trucks or "T-pods." Elevian develops regenerative medicines, with the potential to treat and prevent many age-related diseases and extend healthy lifespan, targeting a fundamental mechanism of aging: regenerative capacity, our body's ability to heal itself, which declines with age.

SAN CARLOS, Calif., Jan. 14, 2019 (GLOBE NEWSWIRE) -- Check Point Software Technologies Ltd. (NASDAQ:CHKP), a leading provider of cyber security solutions globally, today announces it has acquired ForceNock Security Ltd. of Tel Aviv, Israel. Founded in 2017, ForceNock, developed a Web Application and API Protection (WAAP) technology which utilizes machine learning, behavioral and reputation-based security engines. Check Point plans to integrate ForceNock's technology into its Infinity total protection architecture. "Check Point is committed to providing the most comprehensive security architecture to prevent current and future generations of cyber attacks. The growing usage of platforms - Cloud, Network, Mobile, Endpoint and IoT - requires complete, simple to deploy and easy to use security technologies", said Dr. Dorit Dor, Check Point's VP Products. "Incorporating ForceNock's technology into our Infinity Architecture will enable us to continue to provide the highest level of security for our customers worldwide and strengthens our machine learning protection capabilities."

Gender gaps persist in many developed nations and in many fields, and the growing field of artificial intelligence is no exception. In fact, the AI jobs outlook is not looking so great when it comes to gender diversity. Rankings released in December by the World Economic Forum showed that the United States, India, Germany, Switzerland, and Canada have the world's five most concentrated AI workforces. The AI workforce in general has expanded considerably in recent years, with the number of workers with AI skills growing by 190 percent between 2015 and 2017, according to an analysis by LinkedIn. However, the rankings on gender diversity were less than impressive, reflecting an issue that affects the tech industry as a whole but with particular implications for AI and machine learning: a lack of diversity and inclusivity in a workforce that is designing solutions for the future.