Today we're excited to announce the availability of NVIDIA DIGITS 4. DIGITS 4 introduces a new object detection workflow and DetectNet, a new deep neural network for object detection that enables data scientists and researchers to train models that can detect instances of faces, pedestrians, traffic signs, vehicles and other objects in images. Object detection is one of the most challenging problems in computer vision and is the first step in several computer vision applications. The goal of an object detection system is to detect all instances of objects of a known category in an image. Figure 1 shows the final results of an object detection system trained with DIGITS which can detect vehicles on a construction site. Starting with a successful vehicle detection system like this, you can solve a number of other problems such as recognizing the makes and models of the vehicles, counting and tracking vehicle locations over time, generating natural language descriptions of the images and so on.

Kyulux Inc announced that it has a license with Harvard University's Molecular Space Shuttle deep learning system to develop new display and lighting application materials, according to a news release. Kyulux is an advanced materials start-up company that commercializes thermally activated delayed fluorescence (TADF) OLED display and lighting technology. The Molecular Space Shuttle is an artificial intelligence platform designed by Alán Aspuru-Guzik's group at Harvard's chemistry and chemical biology department, where Aspuru-Guzik is a professor.

FANUC, the world's largest maker of industrial robots, plans to start connecting 400,000 of their installed systems by the end of this year. The goal is to collect data about their operations and, through the use of deep learning, improve performance. Similarly, Kuka is building a deep-learning AI network for their industrial robots. FANUC is now moving forward to connect all its manufacturing robots. The system proactively detects and informs of a potential equipment or process problem before unexpected downtime occurs.

Canadian video game star Elliot Carroza-Oyarce won the gold medal Tuesday in the first-ever EGames contest, held amid the Olympics festivities in Rio de Janeiro. Organizers hoped the hours-long "Super Smash Bros." competition would show how the growing number of competitive video gamers across the world could have their own recurring celebration in the years to come. The e-sports community has been divided over whether video games should be in the Olympics, commemorated with their own global championships, or continue as is, with disparate worldwide contests and leagues. Video game players "don't necessarily want to be boxed in with traditional athletes," said Neil Duffy, executive vice president at the British nonprofit behind the EGames. "But the level of skill and practice and effort that this competition requires really is comparable with that of a football or basketball player."

It takes years, decades even, for physicians to refine the expertise required to notice details that remain invisible to the untrained. This aptitude, depending on a doctor's specialty, might mark the difference between an oncologist knowing a malignant tumor from a benign cyst. It can help a cardiologist determine the velocity of blood as it flows through a hole in the heart. Or it may tell a reconstructive plastic surgeon whether a severe burn is healing nicely or at risk of infection. None of this is easy unless you know how to see in a certain way.

Rich and complex time-series data, such as those generated from engineering systems, financial markets, videos or neural recordings, are now a common feature of modern data analysis. Explaining the phenomena underlying these diverse data sets requires flexible and accurate models. In this paper, we promote Gaussian process dynamical systems (GPDS) as a rich model class that is appropriate for such analysis. In particular, we present a message passing algorithm for approximate inference in GPDSs based on expectation propagation. By posing inference as a general message passing problem, we iterate forward-backward smoothing. Thus, we obtain more accurate posterior distributions over latent structures, resulting in improved predictive performance compared to state-of-the-art GPDS smoothers, which are special cases of our general message passing algorithm. Hence, we provide a unifying approach within which to contextualize message passing in GPDSs.

Linear regression, the fitting of linear relationships among variables in a data set, is a standard tool in data analysis. In particular, for the sake of interpretability and utility in further analysis, we desire to find highly sparse linear relationships, i.e., involving only a few variables. Of course, such simple linear relationships often will not hold across an entire population. But, more frequently there will exist conditions - perhaps a range of parameters or a segment of a larger population - under which such sparse models fit the data quite well. For example, Rosenfeld et al. [16] used data mining heuristics to identify small segments of a population in which a few additional risk factors were highly predictive of certain kinds of cancer, whereas these same risk factors were not significant in the overall population. Simple rules for special cases may also hint at the more complex general rules. More generally, we need to develop new techniques to reason about populations in which most members are atypical in some way, which are colloquially (and somewhat abusively) referred to as long-tailed distributions. We are seeking principled alternatives to ad-hoc approaches such as trying a variety of methods for clustering the data and hoping that the identified clusters can be modeled well.

Crop yield forecasting is the methodology of predicting crop yields prior to harvest. The availability of accurate yield prediction frameworks have enormous implications from multiple standpoints, including impact on the crop commodity futures markets, formulation of agricultural policy, as well as crop insurance rating. The focus of this work is to construct a corn yield predictor at the county scale. Corn yield (forecasting) depends on a complex, interconnected set of variables that include economic, agricultural, management and meteorological factors. Conventional forecasting is either knowledge-based computer programs (that simulate plant-weather-soil-management interactions) coupled with targeted surveys or statistical model based. The former is limited by the need for painstaking calibration, while the latter is limited to univariate analysis or similar simplifying assumptions that fail to capture the complex interdependencies affecting yield. In this paper, we propose a data-driven approach that is "gray box" i.e. that seamlessly utilizes expert knowledge in constructing a statistical network model for corn yield forecasting. Our multivariate gray box model is developed on Bayesian network analysis to build a Directed Acyclic Graph (DAG) between predictors and yield. Starting from a complete graph connecting various carefully chosen variables and yield, expert knowledge is used to prune or strengthen edges connecting variables. Subsequently the structure (connectivity and edge weights) of the DAG that maximizes the likelihood of observing the training data is identified via optimization. We curated an extensive set of historical data (1948-2012) for each of the 99 counties in Iowa as data to train the model.

Deep brain stimulation (DBS) is a surgical treatment for Parkinson's Disease. Static models based on quasi-static approximation are common approaches for DBS modeling. While this simplification has been validated for bioelectric sources, its application to rapid stimulation pulses, which contain more high-frequency power, may not be appropriate, as DBS therapeutic results depend on stimulus parameters such as frequency and pulse width, which are related to time variations of the electric field. We propose an alternative hybrid approach based on probabilistic models and differential equations, by using Gaussian processes and wave equation. Our model avoids quasi-static approximation, moreover, it is able to describe dynamic behavior of DBS. Therefore, the proposed model may be used to obtain a more realistic phenomenon description. The proposed model can also solve inverse problems, i.e. to recover the corresponding source of excitation, given electric potential distribution. The electric potential produced by a time-varying source was predicted using proposed model. For static sources, the electric potential produced by different electrode configurations were modeled. Four different sources of excitation were recovered by solving the inverse problem. We compare our outcomes with the electric potential obtained by solving Poisson's equation using the Finite Element Method (FEM). Our approach is able to take into account time variations of the source and the produced field. Also, inverse problem can be addressed using the proposed model. The electric potential calculated with the proposed model is close to the potential obtained by solving Poisson's equation using FEM.

The scale of functional magnetic resonance image data is rapidly increasing as large multi-subject datasets are becoming widely available and high-resolution scanners are adopted. The inherent low-dimensionality of the information in this data has led neuroscientists to consider factor analysis methods to extract and analyze the underlying brain activity. In this work, we consider two recent multi-subject factor analysis methods: the Shared Response Model and Hierarchical Topographic Factor Analysis. We perform analytical, algorithmic, and code optimization to enable multi-node parallel implementations to scale. Single-node improvements result in 99x and 1812x speedups on these two methods, and enables the processing of larger datasets. Our distributed implementations show strong scaling of 3.3x and 5.5x respectively with 20 nodes on real datasets. We also demonstrate weak scaling on a synthetic dataset with 1024 subjects, on up to 1024 nodes and 32,768 cores.