Scientists at the National Center for Supercomputing Applications (NCSA), located at the University of Illinois at Urbana-Champaign, have pioneered the use of GPU-accelerated deep learning for rapid detection and characterization of gravitational waves. This new approach will enable astronomers to study gravitational waves using minimal computational resources, reducing time to discovery and increasing the scientific reach of gravitational wave astrophysics. This innovative research was recently published in Physics Letters B. Combining deep learning algorithms, numerical relativity simulations of black hole mergers - obtained with the Einstein Toolkit run on the Blue Waters supercomputer - and data from the LIGO Open Science Center, NCSA Gravity Group researchers Daniel George and Eliu Huerta produced Deep Filtering, an end-to-end time-series signal processing method. Deep Filtering achieves similar sensitivities and lower errors compared to established gravitational wave detection algorithms, while being far more computationally efficient and more resilient to noise anomalies. The method allows faster than real-time processing of gravitational waves in LIGO's raw data, and also enables new physics, since it can detect new classes of gravitational wave sources that may go unnoticed with existing detection algorithms.

The use of multiple drugs, termed polypharmacy, is common to treat patients with complex diseases or co-existing medical conditions. However, a major consequence of polypharmacy is a much higher risk of side effects for the patient. Polypharmacy side effects emerge because of drug interactions, in which activity of one drug may change, favorably or unfavorably, if taken with another drug. The knowledge of drug interactions is limited because these complex relationships are usually not observed in small clinical testing. Discovering polypharmacy side effects thus remains a challenge with significant implications for patient mortality and morbidity. Here we introduce Decagon, an approach for modeling polypharmacy side effects. The approach constructs a multimodal graph of protein-protein interactions, drug-protein interactions, and the polypharmacy side effects, which are represented as drug-drug interactions, where each side effect is an edge of a different type. Decagon is developed specifically to handle such multimodal graphs with a large number of edge types. Our approach develops a new graph convolutional neural network for multirelational link prediction in multimodal networks. Unlike approaches limited to predicting simple drug-drug interaction values, Decagon can predict the exact side effect, if any, through which a given drug combination manifests clinically. Decagon accurately predicts polypharmacy side effects, outperforming baselines by up to 69%. Furthermore, Decagon models particularly well side effects that have a strong molecular basis, while on predominantly non-molecular side effects, it achieves good performance because of effective sharing of model parameters across edge types. Decagon creates an opportunity to use large molecular and patient population data to flag and prioritize polypharmacy side effects for follow-up analysis via formal pharmacological studies.

Deep neural networks are complex and opaque. As they enter application in a variety of important and safety critical domains, users seek methods to explain their output predictions. We develop an approach to explaining deep neural networks by constructing causal models on salient concepts contained in a CNN. We develop methods to extract salient concepts throughout a target network by using autoencoders trained to extract human-understandable representations of network activations. We then build a bayesian causal model using these extracted concepts as variables in order to explain image classification. Finally, we use this causal model to identify and visualize features with significant causal influence on final classification.

We investigate the automatic classification of patient discharge notes into standard disease labels. We find that Convolutional Neural Networks with Attention outperform previous algorithms used in this task, and suggest further areas for improvement.

We propose an architecture for VQA which utilizes recurrent layers to generate visual and textual attention. The memory characteristic of the proposed recurrent attention units offers a rich joint embedding of visual and textual features and enables the model to reason relations between several parts of the image and question. Our single model outperforms the first place winner on the VQA 1.0 dataset, performs within margin to the current state-of-the-art ensemble model. We also experiment with replacing attention mechanisms in other state-of-the-art models with our implementation and show increased accuracy. In both cases, our recurrent attention mechanism improves performance in tasks requiring sequential or relational reasoning on the VQA dataset.

Though deep neural networks (DNNs) achieve remarkable performances in many artificial intelligence tasks, the lack of training instances remains a notorious challenge. As the network goes deeper, the generalization accuracy decays rapidly in the situation of lacking massive amounts of training data. In this paper, we propose novel deep neural network structures that can be inherited from all existing DNNs with almost the same level of complexity, and develop simple training algorithms. We show our paradigm successfully resolves the lack of data issue. Tests on the CIFAR10 and CIFAR100 image recognition datasets show that the new paradigm leads to 20$\%$ to $30\%$ relative error rate reduction compared to their base DNNs. The intuition of our algorithms for deep residual network stems from theories of the partial differential equation (PDE) control problems. Code will be made available.

Although Recurrent Neural Network (RNN) has been a powerful tool for modeling sequential data, its performance is inadequate when processing sequences with multiple patterns. In this paper, we address this challenge by introducing an external memory and constructing a novel persistent memory augmented RNN (termed as PRNN). The PRNN captures the principle patterns in training sequences and stores them in an external memory. By leveraging the persistent memory, the proposed method can adaptively update states according to the similarities between encoded inputs and memory slots, leading to a stronger capacity in assimilating sequences with multiple patterns. Content-based addressing is suggested in memory accessing, and gradient descent is utilized for implicitly updating the memory. Our approach can be further extended by combining the prior knowledge of data. Experiments on several datasets demonstrate the effectiveness of the proposed method.

Deep residual networks (ResNets) and their variants are widely used in many computer vision applications and natural language processing tasks. However, the theoretical principles for designing and training ResNets are still not fully understood. Recently, several points of view have emerged to try to interpret ResNet theoretically, such as unraveled view, unrolled iterative estimation and dynamical systems view. In this paper, we adopt the dynamical systems point of view, and analyze the lesioning properties of ResNet both theoretically and experimentally. Based on these analyses, we additionally propose a novel method for accelerating ResNet training. We apply the proposed method to train ResNets and Wide ResNets for three image classification benchmarks, reducing training time by more than 40% with superior or on-par accuracy.

Timely accurate traffic forecast is crucial for urban traffic control and guidance. Due to the high nonlinearity and complexity of traffic flow, traditional methods cannot satisfy the requirements of mid-and-long term prediction tasks and often neglect spatial and temporal dependencies. In this paper, we propose a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain. Instead of applying regular convolutional and recurrent units, we formulate the problem on graphs and build the model with complete convolutional structures, which enable much faster training speed with fewer parameters. Experiments show that our STGCN model effectively captures comprehensive spatio-temporal correlations through modeling multi-scale traffic networks and consistently outperforms state-of-the-art baselines on various real-world traffic datasets.

Many practical machine learning tasks employ very deep convolutional neural networks. Such large depths pose formidable computational challenges in training and operating the network. It is therefore important to understand how fast the energy contained in the propagated signals (a.k.a. feature maps) decays across layers. In addition, it is desirable that the feature extractor generated by the network be informative in the sense of the only signal mapping to the all-zeros feature vector being the zero input signal. This "trivial null-set" property can be accomplished by asking for "energy conservation" in the sense of the energy in the feature vector being proportional to that of the corresponding input signal. This paper establishes conditions for energy conservation (and thus for a trivial null-set) for a wide class of deep convolutional neural network-based feature extractors and characterizes corresponding feature map energy decay rates. Specifically, we consider general scattering networks employing the modulus non-linearity and we find that under mild analyticity and high-pass conditions on the filters (which encompass, inter alia, various constructions of Weyl-Heisenberg filters, wavelets, ridgelets, ($\alpha$)-curvelets, and shearlets) the feature map energy decays at least polynomially fast. For broad families of wavelets and Weyl-Heisenberg filters, the guaranteed decay rate is shown to be exponential. Moreover, we provide handy estimates of the number of layers needed to have at least $((1-\varepsilon)\cdot 100)\%$ of the input signal energy be contained in the feature vector.