Given the wide success of convolutional neural networks (CNNs) applied to natural images, researchers have begun to apply them to neuroimaging data. To date, however, exploration of novel CNN architectures tailored to neuroimaging data has been limited. Several recent works fail to leverage the 3D structure of the brain, instead treating the brain as a set of independent 2D slices. Approaches that do utilize 3D convolutions rely on architectures developed for object recognition tasks in natural 2D images. Such architectures make assumptions about the input that may not hold for neuroimaging. For example, existing architectures assume that patterns in the brain exhibit translation invariance. However, a pattern in the brain may have different meaning depending on where in the brain it is located. There is a need to explore novel architectures that are tailored to brain images. We present two simple modifications to existing CNN architectures based on brain image structure. Applied to the task of brain age prediction, our network achieves a mean absolute error (MAE) of 1.4 years and trains 30% faster than a CNN baseline that achieves a MAE of 1.6 years. Our results suggest that lessons learned from developing models on natural images may not directly transfer to neuroimaging tasks. Instead, there remains a large space of unexplored questions regarding model development in this area, whose answers may differ from conventional wisdom.

We propose to use deep neural networks for generating samples in Monte Carlo integration. Our work is based on non-linear independent component analysis, which we extend in numerous ways to improve performance and enable its application to integration problems. First, we introduce piecewise-polynomial coupling transforms that greatly increase the modeling power of individual coupling layers. Second, we propose to preprocess the inputs of neural networks using one-blob encoding, which stimulates localization of computation and improves inference. Third, we derive a gradient-descent-based optimization for the KL and the $\chi^2$ divergence for the specific application of Monte Carlo integration with stochastic estimates of the target distribution. Our approach enables fast and accurate inference and efficient sample generation independent of the dimensionality of the integration domain. We demonstrate the benefits of our approach for generating natural images and in two applications to light-transport simulation. First, we show how to learn joint path-sampling densities in primary sample space and how to importance sample multi-dimensional path prefixes thereof. Second, we use our technique to extract conditional directional densities driven by the triple product of the rendering equation, and leverage them for path guiding. In all applications, our approach yields on-par or higher performance at equal sample count than competing techniques.

Chemical transport models (CTMs), which simulate air pollution transport, transformation, and removal, are computationally expensive, largely because of the computational intensity of the chemical mechanisms: systems of coupled differential equations representing atmospheric chemistry. Here we investigate the potential for machine learning to reproduce the behavior of a chemical mechanism, yet with reduced computational expense. We create a 17-layer residual multi-target regression neural network to emulate the Carbon Bond Mechanism Z (CBM-Z) gas-phase chemical mechanism. We train the network to match CBM-Z predictions of changes in concentrations of 77 chemical species after one hour, given a range of chemical and meteorological input conditions, which it is able to do with root-mean-square error (RMSE) of less than 1.97 ppb (median RMSE = 0.02 ppb), while achieving a 250x computational speedup. An additional 17x speedup (total 4250x speedup) is achieved by running the neural network on a graphics-processing unit (GPU). The neural network is able to reproduce the emergent behavior of the chemical system over diurnal cycles using Euler integration, but additional work is needed to constrain the propagation of errors as simulation time progresses.

In this technical report, we present jLDADMM---an easy-to-use Java toolkit for conventional topic models. jLDADMM is released to provide alternatives for topic modeling on normal or short texts. It provides implementations of the Latent Dirichlet Allocation topic model and the one-topic-per-document Dirichlet Multinomial Mixture model (i.e. mixture of unigrams), using collapsed Gibbs sampling. In addition, jLDADMM supplies a document clustering evaluation to compare topic models. jLDADMM is open-source and available to download at: https://github.com/datquocnguyen/jLDADMM

In the traditional framework of spectral learning of stochastic time series models, model parameters are estimated based on trajectories of fully recorded observations. However, real-world time series data often contain missing values, and worse, the distributions of missingness events over time are often not independent of the visible process. Recently, a spectral OOM learning algorithm for time series with missing data was introduced and proved to be consistent, albeit under quite strong conditions. Here we refine the algorithm and prove that the original strong conditions can be very much relaxed. We validate our theoretical findings by numerical experiments, showing that the algorithm can consistently handle missingness patterns whose dynamic interacts with the visible process.

In this paper we study the limitations of parallelization in convex optimization. A convenient approach to study parallelization is through the prism of \emph{adaptivity} which is an information theoretic measure of the parallel runtime of an algorithm. Informally, adaptivity is the number of sequential rounds an algorithm needs to make when it can execute polynomially-many queries in parallel at every round. For combinatorial optimization with black-box oracle access, the study of adaptivity has recently led to exponential accelerations in parallel runtime and the natural question is whether dramatic accelerations are achievable for convex optimization. Our main result is a spoiler. We show that, in general, parallelization does not accelerate convex optimization. In particular, for the problem of minimizing a non-smooth Lipschitz and strongly convex function with black-box oracle access we give information theoretic lower bounds that indicate that the number of adaptive rounds of any randomized algorithm exactly match the upper bounds of single-query-per-round (i.e. non-parallel) algorithms.

Chinese artificial-intelligence startup CloudWalk Technology signed a deal in March with the Zimbabwean government, providing the authoritarian regime an advanced facial-recognition system that it can use to identify, track, and monitor citizens. In exchange, CloudWalk gains access to the facial data of the demographically distinct country, which provides the company much-needed data for improving its recognition algorithms. Arrangements such as this are common under China's Artificial Intelligence (AI) strategy, whereby Chinese private and state-controlled companies take advantage of the weak legal systems and low privacy standards of developing nations as part of the country's effort to become a world leader in artificial intelligence by 2030. But the vision of artificial intelligence that China is creating is a thoroughly illiberal one. Constant surveillance of citizens is powering initiatives such as the Social Credit System, which will rate citizens on their social and economic performance, increasing the power of the state to enforce its cultural vision.

Recent years of lackluster performance and the most significant profit drop in six years has fast-fashion retailer H&M looking for a road to profitability. The company is turning to tech to build a stronger business, drive efficiencies in its supply chain and operations, and give consumers what they want thanks to the insights provided from big data and artificial intelligence about fashion trends and their customers' preferences. Only time will tell if their investment is enough to catapult them out of their sales slump and if their bet on AI and big data will pay off. Here are a few ways H&M is using tech to their business advantage. About 20 years ago, fast-fashion retailers became disruptors that built strong businesses by trading in quality for better prices and fresh products.

The next Wolfenstein game might not even need to remove Adolf Hitler's mustache: Germany's Entertainment Software Self-Regulation Body (or USK), an independent, industry-funded board that oversees age and content ratings for videos games available in the country, announced on Thursday that it will now permit the sale of games featuring Nazi imagery within the country, something that had previously been banned. The industry body's decision reportedly came after a heated debate involving the Nazi-killing Wolfenstein series, particularly a pair of anti–Third Reich games in 2014 and 2017 that were visibly, and somewhat humorously, self-censored in Germany in order to avoid violating a provision of the country's constitution. Previously, video games with Nazi symbolism were heavily censored or outright banned based on the German criminal code's Section 86a, which forbids the use of symbols, flags, insignia, uniforms, slogans, propaganda, and greetings relating to "unconstitutional organizations" (read: Nazis) in German products. Section 86a violations could be met with up to three years of imprisonment or a hefty fine. While the list of games with German-censored versions is quite long, some of the bigger or more recent affected titles include Wolfenstein: The New Order and Wolfenstein II: The New Colossus.

People who check their phones while eating or spending time with their friends are less likely to enjoy themselves, a study has found. Researchers at the University of British Columbia in Canada found that mobile phone use is making people more distracted, distant and drained as a result of its pervasiveness in our modern lives. Even having a mobile phone within easy access during a meal is enough to make diners not enjoy the experience as much as those who keep their devices out of reach while they eat. The I.F.O. is fuelled by eight electric engines, which is able to push the flying object to an estimated top speed of about 120mph. The giant human-like robot bears a striking resemblance to the military robots starring in the movie'Avatar' and is claimed as a world first by its creators from a South Korean robotic company Waseda University's saxophonist robot WAS-5, developed by professor Atsuo Takanishi and Kaptain Rock playing one string light saber guitar perform jam session A man looks at an exhibit entitled'Mimus' a giant industrial robot which has been reprogrammed to interact with humans during a photocall at the new Design Museum in South Kensington, London Electrification Guru Dr. Wolfgang Ziebart talks about the electric Jaguar I-PACE concept SUV before it was unveiled before the Los Angeles Auto Show in Los Angeles, California, U.S The Jaguar I-PACE Concept car is the start of a new era for Jaguar.