Deep neural networks (DNNs) have achieved unprecedented performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. This has caused a recent surge of interest in methods for rendering modern neural systems more interpretable. In this work, we propose to address the interpretability problem in modern DNNs using the rich history of problem descriptions, theories and experimental methods developed by cognitive psychologists to study the human mind. To explore the potential value of these tools, we chose a well-established analysis from developmental psychology that explains how children learn word labels for objects, and applied that analysis to DNNs. Using datasets of stimuli inspired by the original cognitive psychology experiments, we find that state-of-the-art one shot learning models trained on ImageNet exhibit a similar bias to that observed in humans: they prefer to categorize objects according to shape rather than color. The magnitude of this shape bias varies greatly among architecturally identical, but differently seeded models, and even fluctuates within seeds throughout training, despite nearly equivalent classification performance. These results demonstrate the capability of tools from cognitive psychology for exposing hidden computational properties of DNNs, while concurrently providing us with a computational model for human word learning.

Influence propagation is concerned with the diffusion of information (or viruses) from initially influenced (or infected) nodes, called seeds, in a network. Understanding how information propagates in networks has become a central problem in a broad range of fields, such as viral marketing [17], sociology [8, 19, 22], communication [12], epidemiology [20], and social network analysis [23]. One of the most fundamental questions on influence propagation is to estimate the influence, i.e. the expected number of influenced nodes at the end of the propagation given some seeds. Estimating the influence is central to various research problems related to influence propagation, such as the widely-known influence maximization problem -- finding a set of k nodes that maximizes the expected number of influenced nodes. Recent studies in the influence propagation have proposed heuristic algorithms [11, 18, 3, 7, 21] for the influence maximization problem while using Monte Carlo (MC) simulations to approximate the influence. Despite its simplicity, approximating the influence via MC simulations is far from ideal for large networks; in particular, MC may require a large amount of computations in order to stabilize the approximation.

Virtual screening (VS) is widely used during computational drug discovery to reduce costs. Chemogenomics-based virtual screening (CGBVS) can be used to predict new compound-protein interactions (CPIs) from known CPI network data using several methods, including machine learning and data mining. Although CGBVS facilitates highly efficient and accurate CPI prediction, it has poor performance for prediction of new compounds for which CPIs are unknown. The pairwise kernel method (PKM) is a state-of-the-art CGBVS method and shows high accuracy for prediction of new compounds. In this study, on the basis of link mining, we improved the PKM by combining link indicator kernel (LIK) and chemical similarity and evaluated the accuracy of these methods. The proposed method obtained an average area under the precision-recall curve (AUPR) value of 0.562, which was higher than that achieved by the conventional Gaussian interaction profile (GIP) method (0.425), and the calculation time was only increased by a few percent.

Similarity-based approaches represent a promising direction for time series analysis. However, many such methods rely on parameter tuning, and some have shortcomings if the time series are multivariate (MTS), due to dependencies between attributes, or the time series contain missing data. In this paper, we address these challenges within the powerful context of kernel methods by proposing the robust \emph{time series cluster kernel} (TCK). The approach taken leverages the missing data handling properties of Gaussian mixture models (GMM) augmented with informative prior distributions. An ensemble learning approach is exploited to ensure robustness to parameters by combining the clustering results of many GMM to form the final kernel. We evaluate the TCK on synthetic and real data and compare to other state-of-the-art techniques. The experimental results demonstrate that the TCK is robust to parameter choices, provides competitive results for MTS without missing data and outstanding results for missing data.

This paper presents a unified framework for smooth convex regularization of discrete optimal transport problems. In this context, the regularized optimal transport turns out to be equivalent to a matrix nearness problem with respect to Bregman divergences. Our framework thus naturally generalizes a previously proposed regularization based on the Boltzmann-Shannon entropy related to the Kullback-Leibler divergence, and solved with the Sinkhorn-Knopp algorithm. We call the regularized optimal transport distance the rot mover's distance in reference to the classical earth mover's distance. We develop two generic schemes that we respectively call the alternate scaling algorithm and the non-negative alternate scaling algorithm, to compute efficiently the regularized optimal plans depending on whether the domain of the regularizer lies within the non-negative orthant or not. These schemes are based on Dykstra's algorithm with alternate Bregman projections, and further exploit the Newton-Raphson method when applied to separable divergences. We enhance the separable case with a sparse extension to deal with high data dimensions. We also instantiate our proposed framework and discuss the inherent specificities for well-known regularizers and statistical divergences in the machine learning and information geometry communities. Finally, we demonstrate the merits of our methods with experiments using synthetic data to illustrate the effect of different regularizers and penalties on the solutions, as well as real-world data for a pattern recognition application to audio scene classification.

Amazon's drone delivery program stopped being a joke a while ago, but the company still has to overcome serious challenges to make the technology actually work. One of these is getting drones near enough to large populations so they're more efficient than regular road delivery. Amazon has an idea for that though: Huge.

'The robot at the remote site has different force, humidity and temperature sensors, all capturing information that a doctor would get when they are directly palpating (physically examining) a patient,' explains Professor Angelika Peer, a robotics researcher at the University of the West of England, UK. Prof. Peer is also the project coordinator of the EU-funded ReMeDi project, which is developing the robotic doctor to allow medical professionals to examine patients over huge distances. Through a specially designed surface mounted on a robotic arm, stiffness data of the patient's abdomen is displayed to the human, allowing the doctor to feel what the remote robot feels. This is made possible thanks to a tool called a haptic device, which has a soft surface reminiscent of skin that can recreate the sense of touch through force and changing its shape. During the examination, the doctor sits at a desk facing three screens, one showing the doctor's hand on the faraway patient and a second for teleconferencing with the patient, which will remain an essential part of the exchange.

Looking blindly for matches on Tinder can be tedious, but a new feature for the infamous dating app aims to make things easier. Tinder is testing an even more premium version of its Tinder Plus paid service, called Tinder Gold, that will let you scroll through people who have already liked you. Gold members opening up the app will have the option to jump to a section that shows them a grid of everyone who has already swiped right on their profile. Tinder is testing several price points for its Gold feature. Trials begin in Australia, Argentina, Mexico, and Canada this week.

Editor's note: The TensorFlow API has undergone changes since this series was first published. However, the general ideas are the same, and an otherwise well-structured tutorial such as this provides a great jumping off point and opportunity to consult the API documentation to identify and implement said changes. In this tutorial I'll explain how to build a simple working Recurrent Neural Network in TensorFlow. This is the first in a series of seven parts where various aspects and techniques of building Recurrent Neural Networks in TensorFlow are covered. A short introduction to TensorFlow is available here.

This online course is designed to teach you how to create deep learning Algorithms in Python by two expert Machine Learning & Data Science experts. This course is split into 32 sections which cover over 179 Artificial Neural Network topics using a video format – receive a certificate of completion at the end of the course. Online learning is very flexible (expiry dates may vary from course to course depending on the course provider). Artificial intelligence is growing exponentially. There is no doubt about that.