arXiv subjects

Bin Dong

Publications and source records attributed to Bin Dong.

At least 127 records · Page 7Linked to original sources

arXiv2019

CURE: Curvature Regularization For Missing Data Recovery

Missing data recovery is an important and yet challenging problem in imaging and data science. Successful models often adopt certain carefully chosen regularization. Recently, the low dimension manifold model (LDMM) was introduced by S.Osher et al. and shown effective in image inpainting. They observed that enforcing low dimensionality on image patch manifold serves as a good image regularizer. In this paper, we observe that having only the low dimension manifold regularization is not enough sometimes, and we need smoothness as well. For that, we introduce a new regularization by combining the low dimension manifold regularization with a higher order Curvature Regularization, and we call this new regularization CURE for short. The key step of solving CURE is to solve a biharmonic equation on a manifold. We further introduce a weighted version of CURE, called WeCURE, in a similar manner as the weighted nonlocal Laplacian (WNLL) method. Numerical experiments for image inpainting and semi-supervised learning show that the proposed CURE and WeCURE significantly outperform LDMM and WNLL respectively.

cs.CV↗

arXiv2019

You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle

Deep learning achieves state-of-the-art results in many tasks in computer vision and natural language processing. However, recent works have shown that deep networks can be vulnerable to adversarial perturbations, which raised a serious robustness issue of deep networks. Adversarial training, typically formulated as a robust optimization problem, is an effective way of improving the robustness of deep networks. A major drawback of existing adversarial training algorithms is the computational overhead of the generation of adversarial examples, typically far greater than that of the network training. This leads to the unbearable overall computational cost of adversarial training. In this paper, we show that adversarial training can be cast as a discrete time differential game. Through analyzing the Pontryagin's Maximal Principle (PMP) of the problem, we observe that the adversary update is only coupled with the parameters of the first layer of the network. This inspires us to restrict most of the forward and back propagation within the first layer of the network during adversary updates. This effectively reduces the total number of full forward and backward propagation to only one for each group of adversary updates. Therefore, we refer to this algorithm YOPO (You Only Propagate Once). Numerical experiments demonstrate that YOPO can achieve comparable defense accuracy with approximately 1/5 ~ 1/4 GPU time of the projected gradient descent (PGD) algorithm. Our codes are available at https://https://github.com/a1600012888/YOPO-You-Only-Propagate-Once.

stat.ML↗

arXiv2019

Distillation $\approx$ Early Stopping? Harvesting Dark Knowledge Utilizing Anisotropic Information Retrieval For Overparameterized Neural Network

Distillation is a method to transfer knowledge from one model to another and often achieves higher accuracy with the same capacity. In this paper, we aim to provide a theoretical understanding on what mainly helps with the distillation. Our answer is "early stopping". Assuming that the teacher network is overparameterized, we argue that the teacher network is essentially harvesting dark knowledge from the data via early stopping. This can be justified by a new concept, {Anisotropic Information Retrieval (AIR)}, which means that the neural network tends to fit the informative information first and the non-informative information (including noise) later. Motivated by the recent development on theoretically analyzing overparameterized neural networks, we can characterize AIR by the eigenspace of the Neural Tangent Kernel(NTK). AIR facilities a new understanding of distillation. With that, we further utilize distillation to refine noisy labels. We propose a self-distillation algorithm to sequentially distill knowledge from the network in the previous training epoch to avoid memorizing the wrong labels. We also demonstrate, both theoretically and empirically, that self-distillation can benefit from more than just early stopping. Theoretically, we prove convergence of the proposed algorithm to the ground truth labels for randomly initialized overparameterized neural networks in terms of $\ell_2$ distance, while the previous result was on convergence in $0$-$1$ loss. The theoretical result ensures the learned neural network enjoy a margin on the training data which leads to better generalization. Empirically, we achieve better testing accuracy and entirely avoid early stopping which makes the algorithm more user-friendly.

stat.ML↗

arXiv2019

PDE-Net 2.0: Learning PDEs from Data with A Numeric-Symbolic Hybrid Deep Network

Partial differential equations (PDEs) are commonly derived based on empirical observations. However, recent advances of technology enable us to collect and store massive amount of data, which offers new opportunities for data-driven discovery of PDEs. In this paper, we propose a new deep neural network, called PDE-Net 2.0, to discover (time-dependent) PDEs from observed dynamic data with minor prior knowledge on the underlying mechanism that drives the dynamics. The design of PDE-Net 2.0 is based on our earlier work \cite{Long2018PDE} where the original version of PDE-Net was proposed. PDE-Net 2.0 is a combination of numerical approximation of differential operators by convolutions and a symbolic multi-layer neural network for model recovery. Comparing with existing approaches, PDE-Net 2.0 has the most flexibility and expressive power by learning both differential operators and the nonlinear response function of the underlying PDE model. Numerical experiments show that the PDE-Net 2.0 has the potential to uncover the hidden PDE of the observed dynamics, and predict the dynamical behavior for a relatively long time, even in a noisy environment.

cs.LG↗

arXiv2019

Annotation-Free Cardiac Vessel Segmentation via Knowledge Transfer from Retinal Images

Segmenting coronary arteries is challenging, as classic unsupervised methods fail to produce satisfactory results and modern supervised learning (deep learning) requires manual annotation which is often time-consuming and can some time be infeasible. To solve this problem, we propose a knowledge transfer based shape-consistent generative adversarial network (SC-GAN), which is an annotation-free approach that uses the knowledge from publicly available annotated fundus dataset to segment coronary arteries. The proposed network is trained in an end-to-end fashion, generating and segmenting synthetic images that maintain the background of coronary angiography and preserve the vascular structures of retinal vessels and coronary arteries. We train and evaluate the proposed model on a dataset of 1092 digital subtraction angiography images, and experiments demonstrate the supreme accuracy of the proposed method on coronary arteries segmentation.

eess.IV↗

arXiv2019

Hybrid Integrated Photonics Using Bulk Acoustic Resonators

Microwave frequency acousto-optic modulation is realized by exciting high overtone bulk acoustic wave resonances (HBAR resonances) in the photonic stack. These confined mechanical stress waves transmit exhibit vertically transmitting, high quality factor (Q) acoustic Fabry Perot resonances that extend into the Gigahertz domain, and offer stress-optical interaction with the optical modes of the microresonator. Although HBAR are ubiquitously used in modern communication, and often exploited in superconducting circuits, this is the first time they have been incorporated on a photonic circuit based chip. The electro-acousto-optical interaction observed within the optical modes exhibits high actuation linearity, low actuation power and negligible crosstalk. Using the electro-acousto-optic interaction, fast optical resonance tuning is achieved with sub-nanosecond transduction time. By removing the silicon backreflection, broadband acoustic modulation at 4.1 and 8.7 GHz is realized with a 3 dB bandwidth of 250 MHz each. The novel hybrid HBAR nanophotonic platform demonstrated here, allowing on chip integration of micron-scale acoustic and photonic resonators, can find immediate applications in tunable microwave photonics, high bandwidth soliton microcomb stabilization, compact opto-electronic oscillators, and in microwave to optical conversion schemes. Moreover the hybrid platform allows implementation of momentum biasing, which allows realization of on chip non-reciprocal devices such as isolators or circulators and topological photonic bandstructures.

physics.app-ph↗

arXiv2019

Understanding and Improving Transformer From a Multi-Particle Dynamic System Point of View

The Transformer architecture is widely used in natural language processing. Despite its success, the design principle of the Transformer remains elusive. In this paper, we provide a novel perspective towards understanding the architecture: we show that the Transformer can be mathematically interpreted as a numerical Ordinary Differential Equation (ODE) solver for a convection-diffusion equation in a multi-particle dynamic system. In particular, how words in a sentence are abstracted into contexts by passing through the layers of the Transformer can be interpreted as approximating multiple particles' movement in the space using the Lie-Trotter splitting scheme and the Euler's method. Given this ODE's perspective, the rich literature of numerical analysis can be brought to guide us in designing effective structures beyond the Transformer. As an example, we propose to replace the Lie-Trotter splitting scheme by the Strang-Marchuk splitting scheme, a scheme that is more commonly used and with much lower local truncation errors. The Strang-Marchuk splitting scheme suggests that the self-attention and position-wise feed-forward network (FFN) sub-layers should not be treated equally. Instead, in each layer, two position-wise FFN sub-layers should be used, and the self-attention sub-layer is placed in between. This leads to a brand new architecture. Such an FFN-attention-FFN layer is "Macaron-like", and thus we call the network with this new architecture the Macaron Net. Through extensive experiments, we show that the Macaron Net is superior to the Transformer on both supervised and unsupervised learning tasks. The reproducible codes and pretrained models can be found at https://github.com/zhuohan123/macaron-net

cs.LG↗

arXiv2019

JSR-Net: A Deep Network for Joint Spatial-Radon Domain CT Reconstruction from incomplete data

CT image reconstruction from incomplete data, such as sparse views and limited angle reconstruction, is an important and challenging problem in medical imaging. This work proposes a new deep convolutional neural network (CNN), called JSR-Net, that jointly reconstructs CT images and their associated Radon domain projections. JSR-Net combines the traditional model-based approach with deep architecture design of deep learning. A hybrid loss function is adapted to improve the performance of the JSR-Net making it more effective in protecting important image structures. Numerical experiments demonstrate that JSR-Net outperforms some latest model-based reconstruction methods, as well as a recently proposed deep model.

physics.med-ph↗

arXiv2019

PORT: A piezoelectric optical resonance tuner

This abstract presents an aluminum nitride (AlN) piezoelectric actuator for tuning optical resonance modes of silicon nitride photonic resonators. The AlN actuator is fabricated on top of a thick silicon dioxide cladding that encapsulates the nitride resonator and waveguide. The PORT is defined by undercutting the cladding layer with a lateral silicon etch. It tunes the optical wavelength by 20pm on applying 60 V to the top electrode with a 0.5nA current draw. The thick oxide cladding preserves the resonator's loaded quality factor Q optical of 64,000 across the entire tuning range. The first bending mode is at 1.1MHz enabling a tuning speed of <1 μs.

physics.app-ph↗

arXiv2019

Nostalgic Adam: Weighting more of the past gradients when designing the adaptive learning rate

First-order optimization algorithms have been proven prominent in deep learning. In particular, algorithms such as RMSProp and Adam are extremely popular. However, recent works have pointed out the lack of ``long-term memory" in Adam-like algorithms, which could hamper their performance and lead to divergence. In our study, we observe that there are benefits of weighting more of the past gradients when designing the adaptive learning rate. We therefore propose an algorithm called the Nostalgic Adam (NosAdam) with theoretically guaranteed convergence at the best known convergence rate. NosAdam can be regarded as a fix to the non-convergence issue of Adam in alternative to the recent work of [Reddi et al., 2018]. Our preliminary numerical experiments show that NosAdam is a promising alternative algorithm to Adam. The proofs, code and other supplementary materials can be found in an anonymously shared link.

cs.LG↗

arXiv2018

Whole Brain Susceptibility Mapping Using Harmonic Incompatibility Removal

Quantitative susceptibility mapping (QSM) aims to visualize the three dimensional susceptibility distribution by solving the field-to-source inverse problem using the phase data in magnetic resonance signal. However, the inverse problem is ill-posed since the Fourier transform of integral kernel has zeroes in the frequency domain. Although numerous regularization based models have been proposed to overcome this problem, the incompatibility in the field data has not received enough attention, which leads to deterioration of the recovery. In this paper, we show that the data acquisition process of QSM inherently generates a harmonic incompatibility in the measured local field. Based on such discovery, we propose a novel regularization based susceptibility reconstruction model with an additional sparsity based regularization term on the harmonic incompatibility. Numerical experiments show that the proposed method achieves better performance than the existing approaches.

math.NA↗

arXiv2018

Parallel Transport Convolution: A New Tool for Convolutional Neural Networks on Manifolds

Convolution has been playing a prominent role in various applications in science and engineering for many years. It is the most important operation in convolutional neural networks. There has been a recent growth of interests of research in generalizing convolutions on curved domains such as manifolds and graphs. However, existing approaches cannot preserve all the desirable properties of Euclidean convolutions, namely compactly supported filters, directionality, transferability across different manifolds. In this paper we develop a new generalization of the convolution operation, referred to as parallel transport convolution (PTC), on Riemannian manifolds and their discrete counterparts. PTC is designed based on the parallel transportation which is able to translate information along a manifold and to intrinsically preserve directionality. PTC allows for the construction of compactly supported filters and is also robust to manifold deformations. This enables us to preform wavelet-like operations and to define deep convolutional neural networks on curved domains.

cs.LG↗

arXiv2018

Dynamically Unfolding Recurrent Restorer: A Moving Endpoint Control Method for Image Restoration

In this paper, we propose a new control framework called the moving endpoint control to restore images corrupted by different degradation levels in one model. The proposed control problem contains a restoration dynamics which is modeled by an RNN. The moving endpoint, which is essentially the terminal time of the associated dynamics, is determined by a policy network. We call the proposed model the dynamically unfolding recurrent restorer (DURR). Numerical experiments show that DURR is able to achieve state-of-the-art performances on blind image denoising and JPEG image deblocking. Furthermore, DURR can well generalize to images with higher degradation levels that are not included in the training stage.

cs.CV↗

arXiv2018

End-to-end Lung Nodule Detection in Computed Tomography

Computer aided diagnostic (CAD) system is crucial for modern med-ical imaging. But almost all CAD systems operate on reconstructed images, which were optimized for radiologists. Computer vision can capture features that is subtle to human observers, so it is desirable to design a CAD system op-erating on the raw data. In this paper, we proposed a deep-neural-network-based detection system for lung nodule detection in computed tomography (CT). A primal-dual-type deep reconstruction network was applied first to convert the raw data to the image space, followed by a 3-dimensional convolutional neural network (3D-CNN) for the nodule detection. For efficient network training, the deep reconstruction network and the CNN detector was trained sequentially first, then followed by one epoch of end-to-end fine tuning. The method was evaluated on the Lung Image Database Consortium image collection (LIDC-IDRI) with simulated forward projections. With 144 multi-slice fanbeam pro-jections, the proposed end-to-end detector could achieve comparable sensitivity with the reference detector, which was trained and applied on the fully-sampled image data. It also demonstrated superior detection performance compared to detectors trained on the reconstructed images. The proposed method is general and could be expanded to most detection tasks in medical imaging.

cs.CV↗

arXiv2018

Angular Upsampling in Infant Diffusion MRI Using Neighborhood Matching in x-q Space

Diffusion MRI requires sufficient coverage of the diffusion wavevector space, also known as the q-space, to adequately capture the pattern of water diffusion in various directions and scales. As a result, the acquisition time can be prohibitive for individuals who are unable to stay still in the scanner for an extensive period of time, such as infants. To address this problem, in this paper we harness non-local self-similar information in the x-q space of diffusion MRI data for q-space upsampling. Specifically, we first perform neighborhood matching to establish the relationships of signals in x-q space. The signal relationships are then used to regularize an ill-posed inverse problem related to the estimation of high angular resolution diffusion MRI data from its low-resolution counterpart. Our framework allows information from curved white matter structures to be used for effective regularization of the otherwise ill-posed problem. Extensive evaluations using synthetic and infant diffusion MRI data demonstrate the effectiveness of our method. Compared with the widely adopted interpolation methods using spherical radial basis functions and spherical harmonics, our method is able to produce high angular resolution diffusion MRI data with greater quality, both qualitatively and quantitatively.

physics.med-ph↗

arXiv2018

Distributed Caching for Complex Querying of Raw Arrays

As applications continue to generate multi-dimensional data at exponentially increasing rates, fast analytics to extract meaningful results is becoming extremely important. The database community has developed array databases that alleviate this problem through a series of techniques. In-situ mechanisms provide direct access to raw data in the original format---without loading and partitioning. Parallel processing scales to the largest datasets. In-memory caching reduces latency when the same data are accessed across a workload of queries. However, we are not aware of any work on distributed caching of multi-dimensional raw arrays. In this paper, we introduce a distributed framework for cost-based caching of multi-dimensional arrays in native format. Given a set of files that contain portions of an array and an online query workload, the framework computes an effective caching plan in two stages. First, the plan identifies the cells to be cached locally from each of the input files by continuously refining an evolving R-tree index. In the second stage, an optimal assignment of cells to nodes that collocates dependent cells in order to minimize the overall data transfer is determined. We design cache eviction and placement heuristic algorithms that consider the historical query workload. A thorough experimental evaluation over two real datasets in three file formats confirms the superiority -- by as much as two orders of magnitude -- of the proposed framework over existing techniques in terms of cache overhead and workload execution time.

cs.DB↗

arXiv2018

PDE-Net: Learning PDEs from Data

In this paper, we present an initial attempt to learn evolution PDEs from data. Inspired by the latest development of neural network designs in deep learning, we propose a new feed-forward deep network, called PDE-Net, to fulfill two objectives at the same time: to accurately predict dynamics of complex systems and to uncover the underlying hidden PDE models. The basic idea of the proposed PDE-Net is to learn differential operators by learning convolution kernels (filters), and apply neural networks or other machine learning methods to approximate the unknown nonlinear responses. Comparing with existing approaches, which either assume the form of the nonlinear response is known or fix certain finite difference approximations of differential operators, our approach has the most flexibility by learning both differential operators and the nonlinear responses. A special feature of the proposed PDE-Net is that all filters are properly constrained, which enables us to easily identify the governing PDE models while still maintaining the expressive and predictive power of the network. These constrains are carefully designed by fully exploiting the relation between the orders of differential operators and the orders of sum rules of filters (an important concept originated from wavelet theory). We also discuss relations of the PDE-Net with some existing networks in computer vision such as Network-In-Network (NIN) and Residual Neural Network (ResNet). Numerical experiments show that the PDE-Net has the potential to uncover the hidden PDE of the observed dynamics, and predict the dynamical behavior for a relatively long time, even in a noisy environment.

math.NA↗

arXiv2017

A Re-weighted Joint Spatial-Radon Domain CT Image Reconstruction Model for Metal Artifact Reduction

High density implants such as metals often lead to serious artifacts in the reconstructed CT images which hampers the accuracy of image based diagnosis and treatment planning. In this paper, we propose a novel wavelet frame based CT image reconstruction model to reduce metal artifacts. This model is built on a joint spatial and Radon (projection) domain (JSR) image reconstruction framework with a built-in weighting and re-weighting mechanism in Radon domain to repair degraded projection data. The new weighting strategy used in the proposed model not only makes the regularization in Radon domain by wavelet frame transform more effective, but also makes the commonly assumed linear model for CT imaging a more accurate approximation of the nonlinear physical problem. The proposed model, which will be referred to as the re-weighted JSR model, combines the ideas of the recently proposed wavelet frame based JSR model \cite{Dong2013} and the normalized metal artifact reduction model \cite{meyer2010normalized}, and manages to achieve noticeably better CT reconstruction quality than both methods. To solve the proposed re-weighted JSR model, an efficient alternative iteration algorithm is proposed with guaranteed convergence. Numerical experiments on both simulated and real CT image data demonstrate the effectiveness of the re-weighted JSR model and its advantage over some of the state-of-the-art methods.

physics.med-ph↗