Machine learning has been an emerging tool for various aspects of infectious diseases including tuberculosis surveillance and detection. However, WHO provided no recommendations on using computer-aided tuberculosis detection software because of the small number of studies, methodological limitations, and limited generalizability of the findings. To quantify the generalizability of the machine-learning model, we developed a Deep Convolutional Neural Network (DCNN) model using a TB-specific CXR dataset of one population (National Library of Medicine Shenzhen No.3 Hospital) and tested it with non-TB-specific CXR dataset of another population (National Institute of Health Clinical Centers). The findings suggested that a supervised deep learning model developed by using the training dataset from one population may not have the same diagnostic performance in another population. Technical specification of CXR images, disease severity distribution, overfitting, and overdiagnosis should be examined before implementation in other settings.

We present a numerical framework for approximating unknown governing equations using observation data and deep neural networks (DNN). In particular, we propose to use residual network (ResNet) as the basic building block for equation approximation. We demonstrate that the ResNet block can be considered as a one-step method that is exact in temporal integration. We then present two multi-step methods, recurrent ResNet (RT-ResNet) method and recursive ReNet (RS-ResNet) method. The RT-ResNet is a multi-step method on uniform time steps, whereas the RS-ResNet is an adaptive multi-step method using variable time steps. All three methods presented here are based on integral form of the underlying dynamical system. As a result, they do not require time derivative data for equation recovery and can cope with relatively coarsely distributed trajectory data. Several numerical examples are presented to demonstrate the performance of the methods.

Generative Adversarial Networks (GANs) have shown great results in accurately modeling complex distributions, but their training is known to be difficult due to instabilities caused by a challenging minimax optimization problem. This is especially troublesome given the lack of an evaluation metric that can reliably detect nonconvergent behaviors. We leverage the notion of duality gap from game theory in order to propose a novel convergence metric for GANs that has low computational cost. We verify the validity of the proposed metric for various test scenarios commonly used in the literature. In the past few years, generative models have become extremely popular in the machine learning community. This is largely due to the recent advances in the field of deep learning, which allowed deep neural generators to produce remarkable results for various tasks, including for example image generation (Radford et al., 2015). Two notable approaches in this area are variational auto-encoders (VAEs) (Kingma & Welling, 2013; Rezende et al., 2014), and generative adversarial networks (GAN) (Goodfellow et al., 2014). In this paper, we focus on GANs, which are especially attractive as they circumvent the notoriously hard optimization of the data likelihood and instead use an adversarial game approach for training a generator. Standard GAN approaches aim at finding a pure Nash Equilibrium by using traditional gradientbased techniques to minimize each players cost in an alternating fashion. However, the minimax objective of GANs makes the optimization process challenging.

Gravitational-wave data analysis is rapidly absorbing techniques from deep learning, with a focus on convolutional networks and related methods that treat noisy time series as images. We pursue an alternative approach, in which waveforms are first represented as weighted sums over reduced bases (reduced-order modeling); we then train artificial neural networks to map gravitational-wave source parameters into basis coefficients. Statistical inference proceeds directly in coefficient space, where it is theoretically straightforward and computationally efficient. The neural networks also provide analytic waveform derivatives, which are useful for gradient-based sampling schemes. We demonstrate fast and accurate coefficient interpolation for the case of a four-dimensional binary-inspiral waveform family, and discuss promising applications of our framework in parameter estimation.

Given that South African education is in crisis, strategies for improvement and sustainability of high-quality, up-to-date education must be explored. In the migration of education online, inclusion of machine translation for low-resourced local languages becomes necessary. This paper aims to spur the use of current neural machine translation (NMT) techniques for low-resourced local languages. The paper demonstrates state-of-the-art performance on English-to-Setswana translation using the Autshumato dataset. The use of the Transformer architecture beat previous techniques by 5.33 BLEU points. This demonstrates the promise of using current NMT techniques for African languages.

Deep neural networks, trained with large amount of labeled data, can fail to generalize well when tested with examples from a \emph{target domain} whose distribution differs from the training data distribution, referred as the \emph{source domain}. It can be expensive or even infeasible to obtain required amount of labeled data in all possible domains. Unsupervised domain adaptation sets out to address this problem, aiming to learn a good predictive model for the target domain using labeled examples from the source domain but only unlabeled examples from the target domain. Domain alignment approaches this problem by matching the source and target feature distributions, and has been used as a key component in many state-of-the-art domain adaptation methods. However, matching the marginal feature distributions does not guarantee that the corresponding class conditional distributions will be aligned across the two domains. We propose co-regularized domain alignment for unsupervised domain adaptation, which constructs multiple diverse feature spaces and aligns source and target distributions in each of them individually, while encouraging that alignments agree with each other with regard to the class predictions on the unlabeled target examples. The proposed method is generic and can be used to improve any domain adaptation method which uses domain alignment. We instantiate it in the context of a recent state-of-the-art method and observe that it provides significant performance improvements on several domain adaptation benchmarks.

Operating deep neural networks on devices with limited resources requires the reduction of their memory footprints and computational requirements. In this paper we introduce a training method, called look-up table quantization, LUT-Q, which learns a dictionary and assigns each weight to one of the dictionary's values. We show that this method is very flexible and that many other techniques can be seen as special cases of LUT-Q. For example, we can constrain the dictionary trained with LUT-Q to generate networks with pruned weight matrices or restrict the dictionary to powers-of-two to avoid the need for multiplications. In order to obtain fully multiplier-less networks, we also introduce a multiplier-less version of batch normalization. Extensive experiments on image recognition and object detection tasks show that LUT-Q consistently achieves better performance than other methods with the same quantization bitwidth.

Deep Learning algorithms have recently become the de-facto paradigm for various prediction problems, which include many privacy-preserving applications like online medical image analysis. Presumably, the privacy of data in a deep learning system is a serious concern. There have been several efforts to analyze and exploit the information leakages from deep learning architectures to compromise data privacy. In this paper, however, we attempt to provide an evaluation strategy for such information leakages through deep neural network architectures by considering a case study on Convolutional Neural Network (CNN) based image classifier. The approach takes the aid of low-level hardware information, provided by Hardware Performance Counters (HPCs), during the execution of a CNN classifier and a simple hypothesis testing in order to produce an alarm if there exists any information leakage on the actual input.

We propose a general theoretical method for analyzing the risk bound in the presence of adversaries. In particular, we try to fit the adversarial learning problem into the minimax framework. We first show that the original adversarial learning problem could be reduced to a minimax statistical learning problem by introducing a transport map between distributions. Then we prove a risk bound for this minimax problem in terms of covering numbers. In contrast to previous minimax bounds in \cite{lee,far}, our bound is informative when the radius of the ambiguity set is small. Our method could be applied to multi-class classification problems and commonly-used loss functions such as hinge loss and ramp loss. As two illustrative examples, we derive the adversarial risk bounds for kernel-SVM and deep neural networks. Our results indicate that a stronger adversary might have a negative impact on the complexity of the hypothesis class and the existence of margin could serve as a defense mechanism to counter adversarial attacks.

Origin-destination (OD) matrices are often used in urban planning, where a city is partitioned into regions and an element (i, j) in an OD matrix records the cost (e.g., travel time, fuel consumption, or travel speed) from region i to region j. In this paper, we partition a day into multiple intervals, e.g., 96 15-min intervals and each interval is associated with an OD matrix which represents the costs in the interval; and we consider sparse and stochastic OD matrices, where the elements represent stochastic but not deterministic costs and some elements are missing due to lack of data between two regions. We solve the sparse, stochastic OD matrix forecasting problem. Given a sequence of historical OD matrices that are sparse, we aim at predicting future OD matrices with no empty elements. We propose a generic learning framework to solve the problem by dealing with sparse matrices via matrix factorization and two graph convolutional neural networks and capturing temporal dynamics via recurrent neural network. Empirical studies using two taxi datasets from different countries verify the effectiveness of the proposed framework.