Recently, Convolutional Neural Networks (CNNs) demonstrate a considerable vulnerability to adversarial attacks, which can be easily misled by adversarial perturbations. With more aggressive methods proposed, adversarial attacks can be also applied to the physical world, causing practical issues to various CNN powered applications. Most existing defense works for physical adversarial attacks only focus on eliminating explicit perturbation patterns from inputs, ignoring interpretation and solution to CNN's intrinsic vulnerability. Therefore, most of them depend on considerable data processing costs and lack the expected versatility to different attacks. In this paper, we propose DoPa - a fast and comprehensive CNN defense methodology against physical adversarial attacks. By interpreting the CNN's vulnerability, we find that non-semantic adversarial perturbations can activate CNN with significantly abnormal activations and even overwhelm other semantic input patterns' activations. We improve the CNN recognition process by adding a self-verification stage to analyze the semantics of distinguished activation patterns with only one CNN inference involved. Based on the detection result, we further propose a data recovery methodology to defend the physical adversarial attacks. We apply such detection and defense methodology into both image and audio CNN recognition process. Experiments show that our methodology can achieve an average rate of 90% success for attack detection and 81% accuracy recovery for image physical adversarial attacks. Also, the proposed defense method can achieve a 92% detection successful rate and 77.5% accuracy recovery for audio recognition applications. Moreover, the proposed defense methods are at most 2.3x faster compared to the state-of-the-art defense methods, making them feasible to resource-constrained platforms, such as mobile devices.

Bayesian networks are a central tool in machine learning and artificial intelligence, and make use of conditional independencies to impose structure on joint distributions. However, they are generally not as expressive as deep learning models and inference is hard and slow. In contrast, deep probabilistic models such as sum-product networks (SPNs) capture joint distributions in a tractable fashion, but use little interpretable structure. Here, we extend the notion of SPNs towards conditional distributions, which combine simple conditional models into high-dimensional ones. As shown in our experiments, the resulting conditional SPNs can be naturally used to impose structure on deep probabilistic models, allow for mixed data types, while maintaining fast and efficient inference.

Objective: To evaluate the feasibility of using an attention-based neural network for predicting the risk of readmission within 30 days of discharge from the intensive care unit (ICU) based on longitudinal electronic medical record (EMR) data and to leverage the interpretability of the model to describe patients-at-risk. Methods: A "time-aware attention" model was trained using publicly available EMR data (MIMIC-III) associated with 45,298 ICU stays for 33,150 patients. The analysed EMR data included static (patient demographics) and timestamped variables (diagnoses, procedures, medications, and vital signs). Bayesian inference was used to compute the posterior distribution of network weights. The prediction accuracy of the proposed model was compared with several baseline models and evaluated based on average precision, AUROC, and F1-Score. Odds ratios (ORs) associated with an increased risk of readmission were computed for static variables. Diagnoses, procedures, and medications were ranked according to the associated risk of readmission. The model was also used to generate reports with predicted risk (and associated uncertainty) justified by specific diagnoses, procedures, medications, and vital signs. Results: A Bayesian ensemble of 10 time-aware attention models led to the highest predictive accuracy (average precision: 0.282, AUROC: 0.738, F1-Score: 0.353). Male gender, number of recent admissions, age, admission location, insurance type, and ethnicity were all associated with risk of readmission. A longer length of stay in the ICU was found to reduce the risk of readmission (OR: 0.909, 95% credible interval: 0.902, 0.916). Groups of patients at risk included those requiring cardiovascular or ventilatory support, those with poor nutritional state, and those for whom standard medical care was not suitable, e.g. due to contraindications to surgery or medications.

The signature is an infinite graded sequence of statistics known to characterise a stream of data up to a negligible equivalence class. It is a transform which has previously been treated as a fixed feature transformation, on top of which a model may be built. We propose a novel approach which combines the advantages of the signature transform with modern deep learning frameworks. By learning an augmentation of the stream prior to the signature transform, the terms of the signature may be selected in a data-dependent way. More generally, we describe how the signature transform may be used as a layer anywhere within a neural network. In this context it may be interpreted as an activation function not operating element-wise. We present the results of empirical experiments to back up the theoretical justification.

We study the geometry of deep (neural) networks (DNs) with piecewise affine and convex nonlinearities. The layers of such DNs have been shown to be {\em max-affine spline operators} (MASOs) that partition their input space and apply a region-dependent affine mapping to their input to produce their output. We demonstrate that each MASO layer's input space partitioning corresponds to a {\em power diagram} (an extension of the classical Voronoi tiling) with a number of regions that grows exponentially with respect to the number of units (neurons). We further show that a composition of MASO layers (e.g., the entire DN) produces a progressively subdivided power diagram and provide its analytical form. The subdivision process constrains the affine maps on the (exponentially many) power diagram regions to greatly reduce their complexity. For classification problems, we obtain a formula for a MASO DN's decision boundary in the input space plus a measure of its curvature that depends on the DN's nonlinearities, weights, and architecture. Numerous numerical experiments support and extend our theoretical results.

The increasing accessibility of data provides substantial opportunities for understanding user behaviors. Unearthing anomalies in user behaviors is of particular importance as it helps signal harmful incidents such as network intrusions, terrorist activities, and financial frauds. Many visual analytics methods have been proposed to help understand user behavior-related data in various application domains. In this work, we survey the state of art in visual analytics of anomalous user behaviors and classify them into four categories including social interaction, travel, network communication, and transaction. We further examine the research works in each category in terms of data types, anomaly detection techniques, and visualization techniques, and interaction methods. Finally, we discuss the findings and potential research directions.

D.A., A.P.K., S.B. and B.C. developed the network architecture and data/modeling infrastructure, training and testing setup. D.A. and A.P.K. created the figures, wrote the methods and performed additional analysis requested in the review process. D.P.N. and J.J.R. provided clinical expertise and guidance on the study design. G.C and S.S. advised on the modeling techniques. M.E., S.S., J.J.R., B.C., W.Y. and D.A. created the datasets, interpreted the data and defined the clinical labels.

Deep Learning is really starting to establish itself as a major new tool in visual effects. Currently the tools are still in their infancy but they are changing the way visual effects can be approached. Instead of a pipeline consisting of modelling, texturing, lighting and rendering, these new approaches are hallucinating or plausibly creating imagery that is based on training data sets. Machine Learning, the superset of Deep Learning and similar approaches have had great success in image classification, image recognition and image synthesis. At fxguide we covered Synthesia in the UK, a company born out of research first published as Face2Face.

Hailo, an AI startup based in Israel, has released its initial chip that the company claims is "the world's top performing deep learning processor," with the Hailo-8 chip claimed to deliver 26 tera-operations pers second (TOPS), while consuming only a few watts of power. If true, that would certainly put it near or at the top of its class in performance for edge applications in areas like self-driving cars, drones, smart appliances, and virtual/augmented reality devices. The challenge in these edgey environments has always been to get AI processors with the requisite performance for these applications but consuming only the small amounts of power available in these settings. In fact, Hailo is positioning its new offering as chip that "enables edge devices to run sophisticated deep learning applications that could previously run only on the cloud." However, doesn't mean Hailo-8 is as powerful as a top-of-the-line inference GPU for the datacenter.

Deep Learning has revolutionized the Machine Learning scene in the last years. Can we apply it to image compression? How well can a Deep Learning algorithm reconstruct pictures of kittens? Today we'll find the answers to all of those questions. I've talked about Unsupervised Learning before: applying Machine Learning to discover patterns in unlabelled data.