LAS VEGAS, NV--(Marketwired - Jan 4, 2017) - CES -- NVIDIA (NASDAQ: NVDA) today unveiled the new NVIDIA SHIELD TV -- an Android open-platform media streamer built on bleeding-edge visual computing technology that delivers unmatched experiences in streaming, gaming and AI. Sporting a sleek, new design and now shipping with both a remote and a game controller, SHIELD provides the best, most complete entertainment experience in the living room. "NVIDIA's rich heritage in visual computing and deep learning has enabled us to create this revolutionary device," said Jen-Hsun Huang, founder and chief executive officer of NVIDIA, who revealed SHIELD during his opening keynote address at CES. "SHIELD TV is the world's most advanced streamer. Its brilliant 4K HDR quality, hallmark NVIDIA gaming performance and broad access to media content will bring families hours of joy. And with SHIELD's new AI home capability, we can control and interact with content through the magic of artificial intelligence from anywhere in the house," he said.

Panasonic Corporation of North America announced it has developed a "companion" robot with human-like movements and communication skills. The robot, developed as a proof of concept, debuts at the Panasonic booth (#12908) at CES 2017 in Las Vegas. This Smart News Release features multimedia. "This test project builds on Panasonic's innovations in robotics including battery and power solutions, vision and sensing, navigation solutions and motion control in a new appealing design. This is Panasonic's latest effort in demonstrating network services in a friendly package, and we are showing this robot at CES as a way of obtaining feedback on its features and functions," said Takahiro Iijima, Director, Panasonic Design Strategy Office in North America Director. The robot is equipped with a Wi-Fi network function that accesses artificial intelligence-based natural language processing technology.

BlackBerry QNX is off to a great start at CES 2017. Today they announced what they are calling their "most advanced and secure embedded software platform for autonomous drive and connected cars," which is also known as QNX SDP 7.0. "With the push toward connected and autonomous vehicles, the electronic architecture of cars is evolving – from a multitude of smaller processors each executing a dedicated function, to a set of high performance domain controllers, powered by 64-bit processors and graphical processing units. To develop these new systems, our automotive customers will need a safe and secure 64-bit OS that can run highly complex software, including neural networks and artificial intelligence algorithms. QNX SDP 7.0 is suited not only for cars, but also for almost any safety- or mission-critical application that requires 64-bit performance and advanced security. This includes surgical robots, industrial controllers and high-speed trains."

Delivery driver jobs would be at less risk, CEA forecast says. The proliferation of self-driving, or autonomous, tractor-trailers threaten the jobs of nearly 1.7 million commercial truck drivers, according to a study published late last month by the White House Council of Economic Advisers (CEA). The study, released Dec. 20, said the jobs of between 1.34 million and 1.67 million truck drivers would be at risk due to the growing utilization of heavy-duty vehicles operated via artificial intelligence. That would equal 80 to 100 percent of all driver jobs listed in the CEA report, which is based on May 2015 data from the Bureau of Labor Statistics, a unit of the Department of Labor. There are about 3.4 million commercial truck drivers currently operating in the U.S., according to various estimates.

Bob, CFO of ABC Inc is about to get on an earnings call after just reporting a 20% miss on earnings due to slower revenue growth than forecasted. Company ABC's stock price is plummeting, down 25% in extended hour trading. The board is furious and investors demand answers on the discrepancies. Inaccurate revenue forecast remains one of the biggest risks for CFOs. In a recent study, more than 50% of companies feel their pipeline forecast is only about 50% accurate.

Processes such as disease propagation and information diffusion often spread over some latent network structure which must be learned from observation. Given a set of unlabeled training examples representing occurrences of an event type of interest (e.g., a disease outbreak), our goal is to learn a graph structure that can be used to accurately detect future events of that type. Motivated by new theoretical results on the consistency of constrained and unconstrained subset scans, we propose a novel framework for learning graph structure from unlabeled data by comparing the most anomalous subsets detected with and without the graph constraints. Our framework uses the mean normalized log-likelihood ratio score to measure the quality of a graph structure, and efficiently searches for the highest-scoring graph structure. Using simulated disease outbreaks injected into real-world Emergency Department data from Allegheny County, we show that our method learns a structure similar to the true underlying graph, but enables faster and more accurate detection.

Classical spectral clustering is based on a spectral decomposition of a graph Laplacian, obtained from a graph adjacency matrix representing positive graph edge weights describing similarities of graph vertices. In signed graphs, the graph edge weights can be negative to describe disparities of graph vertices, for example, negative correlations in the data. Negative weights lead to possible negative spectrum of the standard graph Laplacian, which is cured by defining a signed Laplacian. We revisit comparing the standard and signed Laplacians and argue that the former is more natural than the latter, also showing that the negative spectrum is actually beneficial, for spectral clustering of signed graphs.

Computation of moments of transformed random variables is a problem appearing in many engineering applications. The current methods for moment transformation are mostly based on the classical quadrature rules which cannot account for the approximation errors. Our aim is to design a method for moment transformation for Gaussian random variables which accounts for the error in the numerically computed mean. We employ an instance of Bayesian quadrature, called Gaussian process quadrature (GPQ), which allows us to treat the integral itself as a random variable, where the integral variance informs about the incurred integration error. Experiments on the coordinate transformation and nonlinear filtering examples show that the proposed GPQ moment transform performs better than the classical transforms.

In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active towards a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria) it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery.

The problem of outlier detection is extremely challenging in many domains such as text, in which the attribute values are typically non-negative, and most values are zero. In such cases, it often becomes difficult to separate the outliers from the natural variations in the patterns in the underlying data. In this paper, we present a matrix factorization method, which is naturally able to distinguish the anomalies with the use of low rank approximations of the underlying data. Our iterative algorithm TONMF is based on block coordinate descent (BCD) framework. We define blocks over the term-document matrix such that the function becomes solvable. Given most recently updated values of other matrix blocks, we always update one block at a time to its optimal. Our approach has significant advantages over traditional methods for text outlier detection. Finally, we present experimental results illustrating the effectiveness of our method over competing methods.