arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 685 records · Page 38Linked to original sources

arXiv2020

Bringing Old Photos Back to Life

We propose to restore old photos that suffer from severe degradation through a deep learning approach. Unlike conventional restoration tasks that can be solved through supervised learning, the degradation in real photos is complex and the domain gap between synthetic images and real old photos makes the network fail to generalize. Therefore, we propose a novel triplet domain translation network by leveraging real photos along with massive synthetic image pairs. Specifically, we train two variational autoencoders (VAEs) to respectively transform old photos and clean photos into two latent spaces. And the translation between these two latent spaces is learned with synthetic paired data. This translation generalizes well to real photos because the domain gap is closed in the compact latent space. Besides, to address multiple degradations mixed in one old photo, we design a global branch with a partial nonlocal block targeting to the structured defects, such as scratches and dust spots, and a local branch targeting to the unstructured defects, such as noises and blurriness. Two branches are fused in the latent space, leading to improved capability to restore old photos from multiple defects. The proposed method outperforms state-of-the-art methods in terms of visual quality for old photos restoration.

cs.CV↗

arXiv2020

Cross-domain Correspondence Learning for Exemplar-based Image Translation

We present a general framework for exemplar-based image translation, which synthesizes a photo-realistic image from the input in a distinct domain (e.g., semantic segmentation mask, or edge map, or pose keypoints), given an exemplar image. The output has the style (e.g., color, texture) in consistency with the semantically corresponding objects in the exemplar. We propose to jointly learn the crossdomain correspondence and the image translation, where both tasks facilitate each other and thus can be learned with weak supervision. The images from distinct domains are first aligned to an intermediate domain where dense correspondence is established. Then, the network synthesizes images based on the appearance of semantically corresponding patches in the exemplar. We demonstrate the effectiveness of our approach in several image translation tasks. Our method is superior to state-of-the-art methods in terms of image quality significantly, with the image style faithful to the exemplar with semantic consistency. Moreover, we show the utility of our method for several applications

cs.CV↗

arXiv2020

Automatic, Dynamic, and Nearly Optimal Learning Rate Specification by Local Quadratic Approximation

In deep learning tasks, the learning rate determines the update step size in each iteration, which plays a critical role in gradient-based optimization. However, the determination of the appropriate learning rate in practice typically replies on subjective judgement. In this work, we propose a novel optimization method based on local quadratic approximation (LQA). In each update step, given the gradient direction, we locally approximate the loss function by a standard quadratic function of the learning rate. Then, we propose an approximation step to obtain a nearly optimal learning rate in a computationally efficient way. The proposed LQA method has three important features. First, the learning rate is automatically determined in each update step. Second, it is dynamically adjusted according to the current loss function value and the parameter estimates. Third, with the gradient direction fixed, the proposed method leads to nearly the greatest reduction in terms of the loss function. Extensive experiments have been conducted to prove the strengths of the proposed LQA method.

stat.ML↗

arXiv2020

Protocol for an Observational Study on the Effects of Social Distancing on Influenza-Like Illness and COVID-19

The novel coronavirus disease (COVID-19) is a highly contagious respiratory disease that was first detected in Wuhan, China in December 2019, and has since spread around the globe, claiming more than 69,000 lives by the time this protocol is written. It has been widely acknowledged that the most effective public policy to mitigate the pandemic is \emph{social and physical distancing}: keeping at least six feet away from people, working from home, closing non-essential businesses, etc. There have been a lot of anecdotal evidences suggesting that social distancing has a causal effect on disease mitigation; however, few studies have investigated the effect of social distancing on disease mitigation in a transparent and statistically-sound manner. We propose to perform an optimal non-bipartite matching to pair counties with similar observed covariates but vastly different average social distancing scores during the first week (March 16th through Match 22nd) of President's \emph{15 Days to Slow the Spread} campaign. We have produced a total of $302$ pairs of two U.S. counties with good covariate balance on a total of $16$ important variables. Our primary outcome will be the average observed illness collected by Kinsa Inc. two weeks after the intervention period. Although the observed illness does not directly measure COVID-19, it reflects a real-time aspect of the pandemic, and unlike confirmed cases, it is much less confounded by counties' testing capabilities. We also consider observed illness three weeks after the intervention period as a secondary outcome. We will test a proportional treatment effect using a randomization-based test with covariance adjustment and conduct a sensitivity analysis.

q-bio.PE↗

arXiv2020

Sex Differences in Severity and Mortality Among Patients With COVID-19: Evidence from Pooled Literature Analysis and Insights from Integrated Bioinformatic Analysis

Objective: To conduct a meta-analysis of current studies that examined sex differences in severity and mortality in patients with COVID-19, and identify potential mechanisms underpinning these differences. Methods: We performed a systematic review to collate data from observational studies examining associations of sex differences with clinical outcomes of COVID-19. PubMed, Web of Science and four preprint servers were searched for relevant studies. Data were extracted and analyzed using meta-analysis where possible, with summary data presented otherwise. Publicly available bulk RNA sequencing (RNA-seq), single-cell RNA sequencing (scRNA-seq), and chromatin immunoprecipitation sequencing (ChIP-seq) data were analyzed to explore the potential mechanisms underlying the observed association. Results: 39 studies met inclusion criteria, representing 77932 patients, of which 41510 (53.3%) were males. Men were at a markedly increased risk of developing severe cases compared with women. Furthermore, the pooled odds ratio (OR) of mortality for male group compared with the female group indicated significant higher mortality rate for male. Data from scRNA-seq suggest that men have a higher amount of ACE2-expressing pulmonary alveolar type II cells than women. Sex-based immunological differences exist. The expression of androgen receptor (AR) is positively correlated with ACE2, and there is evidence that AR may directly regulate the expression of ACE2. Conclusions: This meta-analysis detected an increased severity and mortality rate in the male populations with COVID-19, which might be attributable to the sex-based differences in cellular compositions and immunological microenvironments of the lung. The host cell receptor ACE2 is likely regulated by AR signaling pathway, which is identified as a potential target for prevention and treatment of SARS-Cov-2 infections in men.

q-bio.PE↗

arXiv2020

Perceptual Image Super-Resolution with Progressive Adversarial Network

Single Image Super-Resolution (SISR) aims to improve resolution of small-size low-quality image from a single one. With popularity of consumer electronics in our daily life, this topic has become more and more attractive. In this paper, we argue that the curse of dimensionality is the underlying reason of limiting the performance of state-of-the-art algorithms. To address this issue, we propose Progressive Adversarial Network (PAN) that is capable of coping with this difficulty for domain-specific image super-resolution. The key principle of PAN is that we do not apply any distance-based reconstruction errors as the loss to be optimized, thus free from the restriction of the curse of dimensionality. To maintain faithful reconstruction precision, we resort to U-Net and progressive growing of neural architecture. The low-level features in encoder can be transferred into decoder to enhance textural details with U-Net. Progressive growing enhances image resolution gradually, thereby preserving precision of recovered image. Moreover, to obtain high-fidelity outputs, we leverage the framework of the powerful StyleGAN to perform adversarial learning. Without the curse of dimensionality, our model can super-resolve large-size images with remarkable photo-realistic details and few distortions. Extensive experiments demonstrate the superiority of our algorithm over state-of-the-arts both quantitatively and qualitatively.

eess.IV↗

arXiv2020

Short-term oscillation and falling dynamics for a water drop dripping in quiescent air

The short-term transient falling dynamics of a dripping water drop in quiescent air has been investigated through both simulation and experiment. The focus is on the short term behavior and the time range considered covers about eight dominant second-mode oscillations of the drop after it is formed. Due to the small fluid inertia the growth of the drop is quasi-static and is well captured by the static pendant drop theory. Nevertheless, the pinching dynamics and the resulting post-formation state of the drop trigger a nonlinear oscillation when the drop falls. The initial shape of the drop when it is just formed is decomposed into spherical harmonic modes. The pinching dynamics such as interface overturning introduces small-scale variation on the drop contour, which in turn contributes to the finite amplitudes of the higher-order modes. Furthermore, the initial kinetic energy when the droplet is just formed is as important as the initial surface energy contained in the drop shape, and is found to amplify the initial oscillation amplitude and to induce a phase shift in the oscillation of all the modes. By incorporating both the initial surface and kinetic energy, the linear model for a free drop oscillation yields very good predictions for the second and third modes. The mode amplitude spectra show both the primary frequencies that are consistent with the Lamb's theory and the secondary frequencies arising from different modes due to nonlinear inter-mode coupling. The complex transient flow inside and outside the drop is induced by the interaction between the falling motion and the nonlinear oscillation. The streamlines indicate that the internal flow is substantially different from the Hill vortex for a falling drop without oscillation. The temporal evolutions of both the internal flow and the wake morphology follow the dominant second oscillation mode.

physics.flu-dyn↗

arXiv2020

MoGA: Searching Beyond MobileNetV3

The evolution of MobileNets has laid a solid foundation for neural network applications on mobile end. With the latest MobileNetV3, neural architecture search again claimed its supremacy in network design. Unfortunately, till today all mobile methods mainly focus on CPU latencies instead of GPU, the latter, however, is much preferred in practice for it has faster speed, lower overhead and less interference. Bearing the target hardware in mind, we propose the first Mobile GPU-Aware (MoGA) neural architecture search in order to be precisely tailored for real-world applications. Further, the ultimate objective to devise a mobile network lies in achieving better performance by maximizing the utilization of bounded resources. Urging higher capability while restraining time consumption is not reconcilable. We alleviate the tension by weighted evolution techniques. Moreover, we encourage increasing the number of parameters for higher representational power. With 200x fewer GPU days than MnasNet, we obtain a series of models that outperform MobileNetV3 under the similar latency constraints, i.e., MoGA-A achieves 75.9% top-1 accuracy on ImageNet, MoGA-B meets 75.5% which costs only 0.5 ms more on mobile GPU. MoGA-C best attests GPU-awareness by reaching 75.3% and being slower on CPU but faster on GPU.The models and test code is made available here https://github.com/xiaomi-automl/MoGA.

cs.LG↗

arXiv2020

Selecting and ranking individualized treatment rules with unmeasured confounding

It is common to compare individualized treatment rules based on the value function, which is the expected potential outcome under the treatment rule. Although the value function is not point-identified when there is unmeasured confounding, it still defines a partial order among the treatment rules under Rosenbaum's sensitivity analysis model. We first consider how to compare two treatment rules with unmeasured confounding in the single-decision setting and then use this pairwise test to rank multiple treatment rules. We consider how to, among many treatment rules, select the best rules and select the rules that are better than a control rule. The proposed methods are illustrated using two real examples, one about the benefit of malaria prevention programs to different age groups and another about the effect of late retirement on senior health in different gender and occupation groups.

stat.ME↗

arXiv2020

A Wasserstein Minimum Velocity Approach to Learning Unnormalized Models

Score matching provides an effective approach to learning flexible unnormalized models, but its scalability is limited by the need to evaluate a second-order derivative. In this paper, we present a scalable approximation to a general family of learning objectives including score matching, by observing a new connection between these objectives and Wasserstein gradient flows. We present applications with promise in learning neural density estimators on manifolds, and training implicit variational and Wasserstein auto-encoders with a manifold-valued prior.

stat.ML↗

arXiv2020

Latent Variables on Spheres for Autoencoders in High Dimensions

Variational Auto-Encoder (VAE) has been widely applied as a fundamental generative model in machine learning. For complex samples like imagery objects or scenes, however, VAE suffers from the dimensional dilemma between reconstruction precision that needs high-dimensional latent codes and probabilistic inference that favors a low-dimensional latent space. By virtue of high-dimensional geometry, we propose a very simple algorithm, called Spherical Auto-Encoder (SAE), completely different from existing VAEs to address the issue. SAE is in essence the vanilla autoencoder with spherical normalization on the latent space. We analyze the unique characteristics of random variables on spheres in high dimensions and argue that random variables on spheres are agnostic to various prior distributions and data modes when the dimension is sufficiently high. Therefore, SAE can harness a high-dimensional latent space to improve the inference precision of latent codes while maintain the property of stochastic sampling from priors. The experiments on sampling and inference validate our theoretical analysis and the superiority of SAE.

cs.LG↗

arXiv2020

To Relieve Your Headache of Training an MRF, Take AdVIL

We propose a black-box algorithm called {\it Adversarial Variational Inference and Learning} (AdVIL) to perform inference and learning on a general Markov random field (MRF). AdVIL employs two variational distributions to approximately infer the latent variables and estimate the partition function of an MRF, respectively. The two variational distributions provide an estimate of the negative log-likelihood of the MRF as a minimax optimization problem, which is solved by stochastic gradient descent. AdVIL is proven convergent under certain conditions. On one hand, compared with contrastive divergence, AdVIL requires a minimal assumption about the model structure and can deal with a broader family of MRFs. On the other hand, compared with existing black-box methods, AdVIL provides a tighter estimate of the log partition function and achieves much better empirical results.

cs.LG↗

arXiv2020

Imaging of buried obstacles in a two-layered medium with phaseless far-field data

The inverse problem we consider is to reconstruct the location and shape of buried obstacles in the lower half-space of an unbounded two-layered medium in two dimensions from phaseless far-field data. A main difficulty of this problem is that the translation invariance property of the modulus of the far field pattern is unavoidable, which is similar to the homogenous background medium case. Based on the idea of using superpositions of two plane waves with different directions as the incident fields, we first develop a direct imaging method to locate the position of small anomalies and give a theoretical analysis of the algorithm. Then a recursive Newton-type iteration algorithm in frequencies is proposed to reconstruct extended obstacles. Finally, numerical experiments are presented to illustrate the feasibility of our algorithms.

math.NA↗

arXiv2020

Privacy for All: Demystify Vulnerability Disparity of Differential Privacy against Membership Inference Attack

Machine learning algorithms, when applied to sensitive data, pose a potential threat to privacy. A growing body of prior work has demonstrated that membership inference attack (MIA) can disclose specific private information in the training data to an attacker. Meanwhile, the algorithmic fairness of machine learning has increasingly caught attention from both academia and industry. Algorithmic fairness ensures that the machine learning models do not discriminate a particular demographic group of individuals (e.g., black and female people). Given that MIA is indeed a learning model, it raises a serious concern if MIA ``fairly'' treats all groups of individuals equally. In other words, whether a particular group is more vulnerable against MIA than the other groups. This paper examines the algorithmic fairness issue in the context of MIA and its defenses. First, for fairness evaluation, it formalizes the notation of vulnerability disparity (VD) to quantify the difference of MIA treatment on different demographic groups. Second, it evaluates VD on four real-world datasets, and shows that VD indeed exists in these datasets. Third, it examines the impacts of differential privacy, as a defense mechanism of MIA, on VD. The results show that although DP brings significant change on VD, it cannot eliminate VD completely. Therefore, fourth, it designs a new mitigation algorithm named FAIRPICK to reduce VD. An extensive set of experimental results demonstrate that FAIRPICK can effectively reduce VD for both with and without the DP deployment.

cs.CR↗

arXiv2020

Deep residual detection of radio frequency interference for FAST

Radio frequency interference (RFI) detection and excision are key steps in the data-processing pipeline of the Five-hundred-meter Aperture Spherical radio Telescope (FAST). Because of its high sensitivity and large data rate, FAST requires more accurate and efficient RFI flagging methods than its counterparts. In the last decades, approaches based upon artificial intelligence (AI), such as codes using convolutional neural networks (CNNs), have been proposed to identify RFI more reliably and efficiently. However, RFI flagging of FAST data with such methods has often proved to be erroneous, with further manual inspections required. In addition, network construction as well as preparation of training data sets for effective RFI flagging has imposed significant additional workloads. Therefore, rapid deployment and adjustment of AI approaches for different observations is impractical to implement with existing algorithms. To overcome such problems, we propose a model called RFI-Net. With the input of raw data without any processing, RFI-Net can detect RFI automatically, producing corresponding masks without any alteration of the original data. Experiments with RFI-Net using simulated astronomical data show that our model has outperformed existing methods in terms of both precision and recall. Besides, compared with other models, our method can obtain the same relative accuracy with fewer training data, thus reducing the effort and time required to prepare the training data set. Further, the training process of RFI-Net can be accelerated, with overfittings being minimized, compared with other CNN codes. The performance of RFI-Net has also been evaluated with observing data obtained by FAST and the Bleien Observatory. Our results demonstrate the ability of RFI-Net to accurately identify RFI with fine-grained, high-precision masks that required no further modification.

astro-ph.IM↗

arXiv2019

Realization of spatial sparseness by deep ReLU nets with massive data

The great success of deep learning poses urgent challenges for understanding its working mechanism and rationality. The depth, structure, and massive size of the data are recognized to be three key ingredients for deep learning. Most of the recent theoretical studies for deep learning focus on the necessity and advantages of depth and structures of neural networks. In this paper, we aim at rigorous verification of the importance of massive data in embodying the out-performance of deep learning. To approximate and learn spatially sparse and smooth functions, we establish a novel sampling theorem in learning theory to show the necessity of massive data. We then prove that implementing the classical empirical risk minimization on some deep nets facilitates in realization of the optimal learning rates derived in the sampling theorem. This perhaps explains why deep learning performs so well in the era of big data.

cs.LG↗

arXiv2019

Automatic quality assessment for 2D fetal sonographic standard plane based on multi-task learning

The quality control of fetal sonographic (FS) images is essential for the correct biometric measurements and fetal anomaly diagnosis. However, quality control requires professional sonographers to perform and is often labor-intensive. To solve this problem, we propose an automatic image quality assessment scheme based on multi-task learning to assist in FS image quality control. An essential criterion for FS image quality control is that all the essential anatomical structures in the section should appear full and remarkable with a clear boundary. Therefore, our scheme aims to identify those essential anatomical structures to judge whether an FS image is the standard image, which is achieved by three convolutional neural networks. The Feature Extraction Network aims to extract deep level features of FS images. Based on the extracted features, the Class Prediction Network determines whether the structure meets the standard and Region Proposal Network identifies its position. The scheme has been applied to three types of fetal sections, which are the head, abdominal, and heart. The experimental results show that our method can make a quality assessment of an FS image within less a second. Also, our method achieves competitive performance in both the detection and classification compared with state-of-the-art methods.

cs.CV↗

arXiv2019

In Vitro Fertilization (IVF) Cumulative Pregnancy Rate Prediction from Basic Patient Characteristics

Tens of millions of women suffer from infertility worldwide each year. In vitro fertilization (IVF) is the best choice for many such patients. However, IVF is expensive, time-consuming, and both physically and emotionally demanding. The first question that a patient usually asks before the IVF is how likely she will conceive, given her basic medical examination information. This paper proposes three approaches to predict the cumulative pregnancy rate after multiple oocyte pickup cycles. Experiments on 11,190 patients showed that first clustering the patients into different groups and then building a support vector machine model for each group can achieve the best overall performance. Our model could be a quick and economic approach for reliably estimating the cumulative pregnancy rate for a patient, given only her basic medical examination information, well before starting the actual IVF procedure. The predictions can help the patient make optimal decisions on whether to use her own oocyte or donor oocyte, how many oocyte pickup cycles she may need, whether to use embryo frozen, etc. They will also reduce the patient's cost and time to pregnancy, and improve her quality of life.

q-bio.QM↗