Video sequences contain rich dynamic patterns, such as dynamic texture patterns that exhibit stationarity in the temporal domain, and action patterns that are non-stationary in either spatial or temporal domain. We show that a spatial-temporal generative ConvNet can be used to model and synthesize dynamic patterns. The model defines a probability distribution on the video sequence, and the log probability is defined by a spatial-temporal ConvNet that consists of multiple layers of spatial-temporal filters to capture spatial-temporal patterns of different scales. The model can be learned from the training video sequences by an "analysis by synthesis" learning algorithm that iterates the following two steps. Step 1 synthesizes video sequences from the currently learned model. Step 2 then updates the model parameters based on the difference between the synthesized video sequences and the observed training sequences. We show that the learning algorithm can synthesize realistic dynamic patterns.

Neural networks have been widely used as predictive models to fit data distribution, and they could be implemented through learning a collection of samples. In many applications, however, the given dataset may contain noisy samples or outliers which may result in a poor learner model in terms of generalization. This paper contributes to a development of robust stochastic configuration networks (RSCNs) for resolving uncertain data regression problems. RSCNs are built on original stochastic configuration networks with weighted least squares method for evaluating the output weights, and the input weights and biases are incrementally and randomly generated by satisfying with a set of inequality constrains. The kernel density estimation (KDE) method is employed to set the penalty weights for each training samples, so that some negative impacts, caused by noisy data or outliers, on the resulting learner model can be reduced. The alternating optimization technique is applied for updating a RSCN model with improved penalty weights computed from the kernel density estimation function. Performance evaluation is carried out by a function approximation, four benchmark datasets and a case study on engineering application. Comparisons to other robust randomised neural modelling techniques, including the probabilistic robust learning algorithm for neural networks with random weights and improved RVFL networks, indicate that the proposed RSCNs with KDE perform favourably and demonstrate good potential for real-world applications.

Neural embeddings have been used with great success in Natural Language Processing (NLP). They provide compact representations that encapsulate word similarity and attain state-of-the-art performance in a range of linguistic tasks. The success of neural embeddings has prompted significant amounts of research into applications in domains other than language. One such domain is graph-structured data, where embeddings of vertices can be learned that encapsulate vertex similarity and improve performance on tasks including edge prediction and vertex labelling. For both NLP and graph based tasks, embeddings have been learned in high-dimensional Euclidean spaces. However, recent work has shown that the appropriate isometric space for embedding complex networks is not the flat Euclidean space, but negatively curved, hyperbolic space. We present a new concept that exploits these recent insights and propose learning neural embeddings of graphs in hyperbolic space. We provide experimental evidence that embedding graphs in their natural geometry significantly improves performance on downstream tasks for several real-world public datasets.

Coresets are small sets of points that approximate the properties of a larger point-set. For example, given a compact set $\mathcal{S} \subseteq \mathbb{R}^d$, a coreset could be defined as a (weighted) subset of $\mathcal{S}$ that approximates the sum of squared distances from $\mathcal{S}$ to every linear subspace of $\mathbb{R}^d$. As such, coresets can be used as a proxy to the full dataset and provide an important technique to speed up algorithms for solving problems including principal component analysis, latent semantic indexing, etc. In this paper, we provide a structural result that connects the construction of such coresets to approximating matrix products. This structural result implies a simple, randomized algorithm that constructs coresets whose sizes are independent of the number and dimensionality of the input points. The expected size of the resulting coresets yields an improvement over the state-of-the-art deterministic approach. Finally, we evaluate the proposed randomized algorithm on synthetic and real data, and demonstrate its effective performance relative to its deterministic counterpart.

We show how the discovery of robust scalable numerical solvers for arbitrary bounded linear operators can be automated as a Game Theory problem by reformulating the process of computing with partial information and limited resources as that of playing underlying hierarchies of adversarial information games. When the solution space is a Banach space $B$ endowed with a quadratic norm $\|\cdot\|$, the optimal measure (mixed strategy) for such games (e.g. the adversarial recovery of $u\in B$, given partial measurements $[\phi_i, u]$ with $\phi_i\in B^*$, using relative error in $\|\cdot\|$-norm as a loss) is a centered Gaussian field $\xi$ solely determined by the norm $\|\cdot\|$, whose conditioning (on measurements) produces optimal bets. When measurements are hierarchical, the process of conditioning this Gaussian field produces a hierarchy of elementary bets (gamblets). These gamblets generalize the notion of Wavelets and Wannier functions in the sense that they are adapted to the norm $\|\cdot\|$ and induce a multi-resolution decomposition of $B$ that is adapted to the eigensubspaces of the operator defining the norm $\|\cdot\|$. When the operator is localized, we show that the resulting gamblets are localized both in space and frequency and introduce the Fast Gamblet Transform (FGT) with rigorous accuracy and (near-linear) complexity estimates. As the FFT can be used to solve and diagonalize arbitrary PDEs with constant coefficients, the FGT can be used to decompose a wide range of continuous linear operators (including arbitrary continuous linear bijections from $H^s_0$ to $H^{-s}$ or to $L^2$) into a sequence of independent linear systems with uniformly bounded condition numbers and leads to $\mathcal{O}(N \operatorname{polylog} N)$ solvers and eigenspace adapted Multiresolution Analysis (resulting in near linear complexity approximation of all eigensubspaces).

Bayesian Neural Networks (BNNs) have recently received increasing attention for their ability to provide well-calibrated posterior uncertainties. However, model selection---even choosing the number of nodes---remains an open question. In this work, we apply a horseshoe prior over node pre-activations of a Bayesian neural network, which effectively turns off nodes that do not help explain the data. We demonstrate that our prior prevents the BNN from under-fitting even when the number of nodes required is grossly over-estimated. Moreover, this model selection over the number of nodes doesn't come at the expense of predictive or computational performance; in fact, we learn smaller networks with comparable predictive performance to current approaches.

Summary: This new study claims to be able to identify criminals based on their facial characteristics. Even if the data science is good has AI pushed too far into areas of societal taboos? This isn't the first time data science has been restricted in favor of social goals, but this study may be a trip wire that starts a long and difficult discussion about the role of AI. Has AI gone too far? This might seem like a nonsensical question to data scientists who strive every day to expand the capabilities of AI until you read the headlines created by this just released peer reviewed scientific paper: Automated Inference on Criminality Using Face Images (Xiaolin Wu, McMaster Univ.

The recent explosion of interest in data science, data mining, and related disciplines has been mirrored by an explosion in book titles on these same topics. This post details the 10 most popular titles in Amazon's Data Mining Books category as of Nov 10, 2016, skipping over repeated titles as well as titles which have been obviously miscategorized and are of no use to our readers. If you want to ask better questions of data, or need to improve and extend the capabilities of your machine learning systems, this practical data science book is invaluable. You'll learn about recent changes to Hadoop, and explore new case studies on Hadoop's role in healthcare systems and genomics data processing.

Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Garry Kasparov, one of the greatest chess players of all time, is famous for his pair of faceoffs against the IBM supercomputer Deep Blue. Kasparov won the first match against the computer, 4-2, in 1996, but lost in the rematch, 3½-2½, in 1997. He recently published a book, "Deep Thinking," about the experience. Business Insider recently spoke with Kasparov about Deep Blue, his thoughts on AI, and machine advancements over the past 20 years -- and how he sees the interplay between machine intelligence and humanity. This interview has been edited for clarity and length. Garry Kasparov: AI as a concept is surrounded by mythology. Most of the things we mention we understand. You know, if we say "white," we all see it's white. If we talk about elements of computer science or some general items, we are in agreement.