arXiv subjects

Jie Zhang

Publications and source records attributed to Jie Zhang.

At least 1,009 records · Page 56Linked to original sources

arXiv2019

Self-supervised Scale Equivariant Network for Weakly Supervised Semantic Segmentation

Weakly supervised semantic segmentation has attracted much research interest in recent years considering its advantage of low labeling cost. Most of the advanced algorithms follow the design principle that expands and constrains the seed regions from class activation maps (CAM). As well-known, conventional CAM tends to be incomplete or over-activated due to weak supervision. Fortunately, we find that semantic segmentation has a characteristic of spatial transformation equivariance, which can form a few self-supervisions to help weakly supervised learning. This work mainly explores the advantages of scale equivariant constrains for CAM generation, formulated as a self-supervised scale equivariant network (SSENet). Specifically, a novel scale equivariant regularization is elaborately designed to ensure consistency of CAMs from the same input image with different resolutions. This novel scale equivariant regularization can guide the whole network to learn more accurate class activation. This regularized CAM can be embedded in most recent advanced weakly supervised semantic segmentation framework. Extensive experiments on PASCAL VOC 2012 datasets demonstrate that our method achieves the state-of-the-art performance both quantitatively and qualitatively for weakly supervised semantic segmentation. Code has been made available.

cs.CV↗

arXiv2019

Video Affective Effects Prediction with Multi-modal Fusion and Shot-Long Temporal Context

Predicting the emotional impact of videos using machine learning is a challenging task considering the varieties of modalities, the complicated temporal contex of the video as well as the time dependency of the emotional states. Feature extraction, multi-modal fusion and temporal context fusion are crucial stages for predicting valence and arousal values in the emotional impact, but have not been successfully exploited. In this paper, we propose a comprehensive framework with novel designs of modal structure and multi-modal fusion strategy. We select the most suitable modalities for valence and arousal tasks respectively and each modal feature is extracted using the modality-specific pre-trained deep model on large generic dataset. Two-time-scale structures, one for the intra-clip and the other for the inter-clip, are proposed to capture the temporal dependency of video content and emotion states. To combine the complementary information from multiple modalities, an effective and efficient residual-based progressive training strategy is proposed. Each modality is step-wisely combined into the multi-modal model, responsible for completing the missing parts of features. With all those improvements above, our proposed prediction framework achieves better performance on the LIRIS-ACCEDE dataset with a large margin compared to the state-of-the-art.

cs.CV↗

arXiv2019

Generative Visual Dialogue System via Adaptive Reasoning and Weighted Likelihood Estimation

The key challenge of generative Visual Dialogue (VD) systems is to respond to human queries with informative answers in natural and contiguous conversation flow. Traditional Maximum Likelihood Estimation (MLE)-based methods only learn from positive responses but ignore the negative responses, and consequently tend to yield safe or generic responses. To address this issue, we propose a novel training scheme in conjunction with weighted likelihood estimation (WLE) method. Furthermore, an adaptive multi-modal reasoning module is designed, to accommodate various dialogue scenarios automatically and select relevant information accordingly. The experimental results on the VisDial benchmark demonstrate the superiority of our proposed algorithm over other state-of-the-art approaches, with an improvement of 5.81% on recall@10.

cs.CV↗

arXiv2019

FocusNet: Imbalanced Large and Small Organ Segmentation with an End-to-End Deep Neural Network for Head and Neck CT Images

In this paper, we propose an end-to-end deep neural network for solving the problem of imbalanced large and small organ segmentation in head and neck (HaN) CT images. To conduct radiotherapy planning for nasopharyngeal cancer, more than 10 organs-at-risk (normal organs) need to be precisely segmented in advance. However, the size ratio between large and small organs in the head could reach hundreds. Directly using such imbalanced organ annotations to train deep neural networks generally leads to inaccurate small-organ label maps. We propose a novel end-to-end deep neural network to solve this challenging problem by automatically locating, ROI-pooling, and segmenting small organs with specifically designed small-organ sub-networks while maintaining the accuracy of large organ segmentation. A strong main network with densely connected atrous spatial pyramid pooling and squeeze-and-excitation modules is used for segmenting large organs, where large organs' label maps are directly output. For small organs, their probabilistic locations instead of label maps are estimated by the main network. High-resolution and multi-scale feature volumes for each small organ are ROI-pooled according to their locations and are fed into small-organ networks for accurate segmenting small organs. Our proposed network is extensively tested on both collected real data and the \emph{MICCAI Head and Neck Auto Segmentation Challenge 2015} dataset, and shows superior performance compared with state-of-the-art segmentation methods.

eess.IV↗

arXiv2019

Imbalanced Learning-based Automatic SAR Images Change Detection by Morphologically Supervised PCA-Net

Change detection is a quite challenging task due to the imbalance between unchanged and changed class. In addition, the traditional difference map generated by log-ratio is subject to the speckle, which will reduce the accuracy. In this letter, an imbalanced learning-based change detection is proposed based on PCA network (PCA-Net), where a supervised PCA-Net is designed to obtain the robust features directly from given multitemporal SAR images instead of a difference map. Furthermore, to tackle with the imbalance between changed and unchanged classes, we propose a morphologically supervised learning method, where the knowledge in the pixels near the boundary between two classes are exploited to guide network training. Finally, our proposed PCA-Net can be trained by the datasets with available reference maps and applied to a new dataset, which is quite practical in change detection projects. Our proposed method is verified on five sets of multiple temporal SAR images. It is demonstrated from the experiment results that with the knowledge in training samples from the boundary, the learned features benefit for change detection and make the proposed method outperforms than supervised methods trained by randomly drawing samples.

cs.CV↗

arXiv2019

Femtosecond relativistic electron beam with reduced timing jitter from THz-driven beam compression

We propose and demonstrate a novel method to reduce the pulse width and timing jitter of a relativistic electron beam through THz-driven beam compression. In this method the longitudinal phase space of a relativistic electron beam is manipulated by a linearly polarized THz pulse in a dielectric tube such that the bunch tail has a higher velocity than the bunch head, which allows simultaneous reduction of both pulse width and timing jitter after passing through a drift. In this experiment, the beam is compressed by more than a factor of four from 130 fs to 28 fs with the arrival time jitter also reduced from 97 fs to 36 fs, opening up new opportunities in using pulsed electron beams for studies of ultrafast dynamics. This technique extends the well known rf buncher to the THz frequency and may have a strong impact in accelerator and ultrafast science facilities that require femtosecond electron beams with tight synchronization to external lasers.

physics.acc-ph↗

arXiv2019

Social Cost Guarantees in Smart Route Guidance

We model and study the problem of assigning traffic in an urban road network infrastructure. In our model, each driver submits their intended destination and is assigned a route to follow that minimizes the social cost (i.e., travel distance of all the drivers). We assume drivers are strategic and try to manipulate the system (i.e., misreport their intended destination and/or deviate from the assigned route) if they can reduce their travel distance by doing so. Such strategic behavior is highly undesirable as it can lead to an overall suboptimal traffic assignment and cause congestion. To alleviate this problem, we develop moneyless mechanisms that are resilient to manipulation by the agents and offer provable approximation guarantees on the social cost obtained by the solution. We then empirically test the mechanisms studied in the paper, showing that they can be effectively used in practice in order to compute manipulation resistant traffic allocations.

cs.GT↗

arXiv2019

Exploring Fault-Tolerant Erasure Codes for Scalable All-Flash Array Clusters

Large-scale systems with all-flash arrays have become increasingly common in many computing segments. To make such systems resilient, we can adopt erasure coding such as Reed-Solomon (RS) code as an alternative to replication because erasure coding incurs a significantly lower storage overhead than replication. To understand the impact of using erasure coding on the system performance and other system aspects such as CPU utilization and network traffic, we build a storage cluster that consists of approximately 100 processor cores with more than 50 high-performance solid-state drives (SSDs), and evaluate the cluster with a popular open-source distributed parallel file system, called Ceph. Specifically, we analyze the behaviors of a system adopting erasure coding from the following five viewpoints, and compare with those of another system using replication: (1) storage system I/O performance; (2) computing and software overheads; (3) I/O amplification; (4) network traffic among storage nodes, and (5) impact of physical data layout on performance of RS-coded SSD arrays. For all these analyses, we examine two representative RS configurations, used by Google file systems, and compare them with triple replication employed by a typical parallel file system as a default fault tolerance mechanism. Lastly, we collect 96 block-level traces from the cluster and release them to the public domain for the use of other researchers.

cs.DC↗

arXiv2019

Average-case Analysis of the Assignment Problem with Independent Preferences

The fundamental assignment problem is in search of welfare maximization mechanisms to allocate items to agents when the private preferences over indivisible items are provided by self-interested agents. The mainstream mechanism \textit{Random Priority} is asymptotically the best mechanism for this purpose, when comparing its welfare to the optimal social welfare using the canonical \textit{worst-case approximation ratio}. Despite its popularity, the efficiency loss indicated by the worst-case ratio does not have a constant bound. Recently, [Deng, Gao, Zhang 2017] show that when the agents' preferences are drawn from a uniform distribution, its \textit{average-case approximation ratio} is upper bounded by 3.718. They left it as an open question of whether a constant ratio holds for general scenarios. In this paper, we offer an affirmative answer to this question by showing that the ratio is bounded by $1/μ$ when the preference values are independent and identically distributed random variables, where $μ$ is the expectation of the value distribution. This upper bound also improves the upper bound of 3.718 in [Deng, Gao, Zhang 2017] for the Uniform distribution. Moreover, under mild conditions, the ratio has a \textit{constant} bound for any independent random values. En route to these results, we develop powerful tools to show the insights that in most instances the efficiency loss is small.

cs.GT↗

arXiv2019

Isovector scalar $a_0(980)$ and $a_0(1450)$ resonances in the $B\rightarrow ψ(K\bar{K},πη) $ decays

We present an analysis of two isovector scalar resonant contributions to the $B$ decays into charmonia plus $K\bar K$ or $πη$ pair in the perturbative QCD approach. The Flatté model for the $a_0(980)$ resonance and the Breit Wigner formula for the $a_0(1450)$ resonance are adopted to parametrize the timelike form factors in the dimeson distribution amplitudes, which capture the important final state interactions in these processes. The predicted distribution in the $K^+K^-$ invariant mass as well as its integrated branching ratio for the $a_0(980)$ resonance in the $B^0\rightarrow J/ψK^+K^-$ mode agree well with the current available experimental data. The obtained branching ratio of the quasi-two-body decay $B^0\rightarrow J/ψa_0(980)(\rightarrow π^0η)$ can reach the order of $10^{-6}$, letting the corresponding measurement appear feasible. For the $a_0(1450)$ component, our results could be tested by further experiments in the LHCb and Belle II. We also discuss some theoretical uncertainties in detail in our calculation.

hep-ph↗

arXiv2019

Why torus-unstable solar filaments experience failed eruption?

To investigate the factors that control the success and/or failure of solar eruptions, we study the magnetic field and 3-Dimensional (3D) configuration of 16 filament eruptions during 2010 July - 2013 February. All these events, i.e., erupted but failed to be ejected to become a coronal mass ejection (CME), are failed eruptions with the filament maximum height exceeding $100 Mm$. The magnetic field of filament source regions is approximated by a potential field extrapolation method. The filament 3D configuration is reconstructed from three vantage points by the observations of STEREO Ahead/Behind and SDO spacecraft. We calculate the decay index at the apex of these failed filaments and find that in 7 cases, their apex decay indexes exceed the theoretical threshold ($n_{crit} = 1.5$) of the torus instability. We further determine the orientation change or rotation angle of each filament top during the eruption. Finally, the distribution of these events in the parameter space of rotation angle versus decay index is established. Four distinct regimes in the parameter space are empirically identified. We find that, all the torus-unstable cases (decay index $n > 1.5$), have a large rotation angles ranging from $50^\circ - 130^\circ$. The possible mechanisms leading to the rotation and failed eruption are discussed. These results imply that, besides the torus instability, the rotation motion during the eruption may also play a significant role in solar eruptions.

astro-ph.SR↗

arXiv2019

Ionic Organic Cage-encapsulating Phase-transferable Metal Clusters

Exploration of metal clusters (MCs) adaptive to both aqueous and oil phases without disturbing their size is promising for a broad scope of applications. The state-of-the-art approach via ligand-binding may perturb MC size due to varied metal-ligand binding strength when shuttling between solvents of different polarity. Herein, we applied physical confinement of a series of small noble MCs (less than 1 nm) inside ionic organic cages (I-Cages), which by means of anion exchange enables reversible transfer of MCs between aqueous and hydrophobic solutions without varying their ultrasmall size. Moreover, the MCs(at)I-Cage hybrid serves as a recyclable, reaction-switchable catalyst featuring high activity in liquid-phase catalysis.

physics.chem-ph↗

arXiv2019

Mass Loss via Solar Wind and Coronal Mass Ejections During Solar Cycle 23 and 24

Similar to the Sun, other stars shed mass and magnetic flux via ubiquitous quasi-steady wind and episodic stellar coronal mass ejections (CMEs). We investigate the mass loss rate via solar wind and CMEs as a function of solar magnetic variability represented in terms of sunspot number and solar X-ray background luminosity. We estimate the contribution of CMEs to the total solar wind mass flux in the ecliptic and beyond, and its variation over different phases of the solar activity cycles. The study exploits the number of sunspots observed, coronagraphic observations of CMEs near the Sun by SOHO/LASCO, in situ observations of the solar wind at 1 AU by WIND, and GOES X-ray flux during solar cycle 23 and 24. We note that the X-ray background luminosity, occurrence rate of CMEs and ICMEs, solar wind mass flux, and associated mass loss rates from the Sun do not decrease as strongly as the sunspot number from the maximum of solar cycle 23 to the next maximum. Our study confirms a true physical increase in CME activity relative to the sunspot number in cycle 24. We show that the CME occurrence rate and associated mass loss rate can be better predicted by X-ray background luminosity than the sunspot number. The solar wind mass loss rate which is an order of magnitude more than the CME mass loss rate shows no obvious dependency on cyclic variation in sunspot number and solar X-ray background luminosity. These results have implications to the study of solar-type stars.

astro-ph.SR↗

arXiv2019

Optimal control of batch processes via a deterministic Q-learning method

Dynamic optimization of nonlinear chemical systems -- such as batch reactors -- should be applied online, and the suitable control taken should be according to the current state of the system rather than the current time instant. The recent state of the art methods applies the control based on the current time instant only. This is not suitable for most cases, as it is not robust to possible changes in the system. This paper proposes a Deterministic Q-Learning method to conduct robust online optimization of batch reactors. In this paper, the Q-Learning method is applied on simple batch reactor models; and in order to show the effectiveness of the proposed method the results are compared to other dynamic optimization methods. The main advantage of the Q-learning method or the proposed method is that it can accommodate unplanned changes during the process via changing the control action; i.e. the main advantage of the proposed method that it can overcome sudden changes during the reaction. In general, we try to maximize the final product obtained or meet certain specifications of the products (e.g. minimize side products).

eess.SY↗

arXiv2019

Regularize, Expand and Compress: Multi-task based Lifelong Learning via NonExpansive AutoML

Lifelong learning, the problem of continual learning where tasks arrive in sequence, has been lately attracting more attention in the computer vision community. The aim of lifelong learning is to develop a system that can learn new tasks while maintaining the performance on the previously learned tasks. However, there are two obstacles for lifelong learning of deep neural networks: catastrophic forgetting and capacity limitation. To solve the above issues, inspired by the recent breakthroughs in automatically learning good neural network architectures, we develop a Multi-task based lifelong learning via nonexpansive AutoML framework termed Regularize, Expand and Compress (REC). REC is composed of three stages: 1) continually learns the sequential tasks without the learned tasks' data via a newly proposed multi-task weight consolidation (MWC) algorithm; 2) expands the network to help the lifelong learning with potentially improved model capability and performance by network-transformation based AutoML; 3) compresses the expanded model after learning every new task to maintain model efficiency and performance. The proposed MWC and REC algorithms achieve superior performance over other lifelong learning algorithms on four different datasets.

cs.CV↗

arXiv2019

FUSE: Fusing STT-MRAM into GPUs to Alleviate Off-Chip Memory Access Overheads

In this work, we propose FUSE, a novel GPU cache system that integrates spin-transfer torque magnetic random-access memory (STT-MRAM) into the on-chip L1D cache. FUSE can minimize the number of outgoing memory accesses over the interconnection network of GPU's multiprocessors, which in turn can considerably improve the level of massive computing parallelism in GPUs. Specifically, FUSE predicts a read-level of GPU memory accesses by extracting GPU runtime information and places write-once-read-multiple (WORM) data blocks into the STT-MRAM, while accommodating write-multiple data blocks over a small portion of SRAM in the L1D cache. To further reduce the off-chip memory accesses, FUSE also allows WORM data blocks to be allocated anywhere in the STT-MRAM by approximating the associativity with the limited number of tag comparators and I/O peripherals. Our evaluation results show that, in comparison to a traditional GPU cache, our proposed heterogeneous cache reduces the number of outgoing memory references by 32% across the interconnection network, thereby improving the overall performance by 217% and reducing energy cost by 53%.

cs.AR↗

arXiv2019

Flux density measurements for 32 pulsars in the 20 cm band

Flux density measurements provide fundamental observational parameters that describe a pulsar. In the current pulsar catalogue, 27% of radio pulsars have no flux density measurement in the 20 cm observing band. Here, we present the first measurements of the flux densities in this band for 32 pulsars observed using the Parkes radio telescope and provide updated pulse profiles for these pulsars. We have used both archival and new observations to make these measurements. Various schemes exist for measuring flux densities. We show how the flux densities measured vary between these methods and how the presence of radio-frequency-interference will bias flux density measurements

astro-ph.GA↗

arXiv2019

Advanced Denoising for X-ray Ptychography

The success of ptychographic imaging experiments strongly depends on achieving high signal-to-noise ratio. This is particularly important in nanoscale imaging experiments when diffraction signals are very weak and the experiments are accompanied by significant parasitic scattering (background), outliers or correlated noise sources. It is also critical when rare events such as cosmic rays, or bad frames caused by electronic glitches or shutter timing malfunction take place. In this paper, we propose a novel iterative algorithm with rigorous analysis that exploits the direct forward model for parasitic noise and sample smoothness to achieve a thorough characterization and removal of structured and random noise. We present a formal description of the proposed algorithm and prove its convergence under mild conditions. Numerical experiments from simulations and real data (both soft and hard X-ray beamlines) demonstrate that the proposed algorithms produce better results when compared to state-of-the-art methods.

eess.IV↗