arXiv subjects

Qian Wang

Publications and source records attributed to Qian Wang.

At least 469 records · Page 26Linked to original sources

arXiv2020

Data Augmentation with norm-VAE for Unsupervised Domain Adaptation

We address the Unsupervised Domain Adaptation (UDA) problem in image classification from a new perspective. In contrast to most existing works which either align the data distributions or learn domain-invariant features, we directly learn a unified classifier for both domains within a high-dimensional homogeneous feature space without explicit domain adaptation. To this end, we employ the effective Selective Pseudo-Labelling (SPL) techniques to take advantage of the unlabelled samples in the target domain. Surprisingly, data distribution discrepancy across the source and target domains can be well handled by a computationally simple classifier (e.g., a shallow Multi-Layer Perceptron) trained in the original feature space. Besides, we propose a novel generative model norm-VAE to generate synthetic features for the target domain as a data augmentation strategy to enhance classifier training. Experimental results on several benchmark datasets demonstrate the pseudo-labelling strategy itself can lead to comparable performance to many state-of-the-art methods whilst the use of norm-VAE for feature augmentation can further improve the performance in most cases. As a result, our proposed methods (i.e. naive-SPL and norm-VAE-SPL) can achieve new state-of-the-art performance with the average accuracy of 93.4% and 90.4% on Office-Caltech and ImageCLEF-DA datasets, and comparable performance on Digits, Office31 and Office-Home datasets with the average accuracy of 97.2%, 87.6% and 67.9% respectively.

cs.CV↗

arXiv2020

Signatures of excited state quantum phase transitions in quantum many body systems: Phase space analysis

Using the Husimi function, we investigate the phase space signatures of the excited state quantum phase transitions (ESQPTs) in the Lipkin and coupled top models. We show that the time evolution of the Husimi function exhibits distinct behaviors between the different phases of an ESQPT and the presence of an ESQPT is signaled by the particular dynamics of the Husimi function. We also evaluate the long time averaged Husimi function and its associated marginal distributions, and discuss how to identify the signatures of ESQPT from their properties. Moreover, by exploiting the second moment and Wherl entropy of the long-time averaged Husimi function, we estimate the critical points of ESQPTs, demonstrating a good agreement with the analytical results. We thus provide further evidence that the phase space methods is a valuable tool for the studies of phase transitions and also open a new way to detect ESQPTs.

quant-ph↗

arXiv2020

The $X_0(2900)$ and its heavy quark spin partners in molecular picture

The $X_0(2900)$ observed by the LHCb Collaboration recently in the $D^-K^+$ invariant mass of the $B^+\to D^+D^-K^+$ process is the first exotic candidate with four different flavors, which opens a new era for the hadron community. Under the assumption that the $X_0(2900)$ is a $I(J^P)=0(0^+)$ $\bar{D}^*K^*$ hadronic molecule, we extract the whole heavy-quark symmetry multiplet formed by the $\left(\bar{D},\bar{D}^*\right)$ doublet and the $K^*$ meson. For the bound state case, there would be two additional $I(J^P)=0(1^+)$ hadronic molecules associated with the $\bar{D}K^*$ and $\bar{D}^*K^*$ channels as well as one additional $I(J^P)=0(2^+)$ $\bar{D}^*K^*$ molecule. In the light quark limit, they are $36.66~\mathrm{MeV}$ and $34.22~\mathrm{MeV}$ below the $\bar{D}K^*$ and $\bar{D}^*K^*$ thresholds, respectively, which are unambiguously fixed by the mass position of the $X_0(2900)$. For the virtual state case, there would be one additional $I(J^P)=0(1^+)$ hadronic molecule strongly coupling to the $\bar{D}K^*$ channel and one additional $I(J^P)=0(2^+)$ $\bar{D}^*K^*$ molecule. Searching for these heavy quark spin partners will help shed light on the nature of the $X_0(2900)$.

hep-ph↗

arXiv2020

The nature of charged charmonium-like states $Z_c(3900)$ and its strange partner $Z_{cs}(3982)$

The observation of $Z_{cs}(3985)$ by BESIII in the $D_s^{*-}D^0$ and $D_s^{-}D^{*0}$ channel adds important dynamic information for a coherent understanding of both $Z_c(3900)$ and $Z_c(4020)$, as well as their strange partners $Z_{cs}(3985)$ and $Z_{cs}(4120)$, on the basis of heavy quark spin symmetry. With an overall short-ranged contact potential in SU(3) flavor symmetry we show that the pole structures can be extracted from the experimental data. Meanwhile, we show the key role played by the $S$-wave open thresholds in $e^+e^-$ annihilations which are correlated with the production of these $Z$ states in a hadronic molecule picture. Implications of their scalar heavy quark spin partners $W_{c0,1,2}$ and $W_{cs0,1,2}$ are also gained.

hep-ph↗

arXiv2020

A Self-ensembling Framework for Semi-supervised Knee Cartilage Defects Assessment with Dual-Consistency

Knee osteoarthritis (OA) is one of the most common musculoskeletal disorders and requires early-stage diagnosis. Nowadays, the deep convolutional neural networks have achieved greatly in the computer-aided diagnosis field. However, the construction of the deep learning models usually requires great amounts of annotated data, which is generally high-cost. In this paper, we propose a novel approach for knee cartilage defects assessment, including severity classification and lesion localization. This can be treated as a subtask of knee OA diagnosis. Particularly, we design a self-ensembling framework, which is composed of a student network and a teacher network with the same structure. The student network learns from both labeled data and unlabeled data and the teacher network averages the student model weights through the training course. A novel attention loss function is developed to obtain accurate attention masks. With dual-consistency checking of the attention in the lesion classification and localization, the two networks can gradually optimize the attention distribution and improve the performance of each other, whereas the training relies on partially labeled data only and follows the semi-supervised manner. Experiments show that the proposed method can significantly improve the self-ensembling performance in both knee cartilage defects classification and localization, and also greatly reduce the needs of annotated data.

eess.IV↗

arXiv2020

Generalized logistic growth modeling of the COVID-19 outbreak: comparing the dynamics in the 29 provinces in China and in the rest of the world

Started in Wuhan, China, the COVID-19 has been spreading all over the world. We calibrate the logistic growth model, the generalized logistic growth model, the generalized Richards model and the generalized growth model to the reported number of infected cases for the whole of China, 29 provinces in China, and 33 countries and regions that have been or are undergoing major outbreaks. We dissect the development of the epidemics in China and the impact of the drastic control measures both at the aggregate level and within each province. We quantitatively document four phases of the outbreak in China with a detailed analysis on the heterogeneous situations across provinces. The extreme containment measures implemented by China were very effective with some instructive variations across provinces. Borrowing from the experience of China, we made scenario projections on the development of the outbreak in other countries. We identified that outbreaks in 14 countries (mostly in western Europe) have ended, while resurgences of cases have been identified in several among them. The modeling results clearly show longer after-peak trajectories in western countries, in contrast to most provinces in China where the after-peak trajectory is characterized by a much faster decay. We identified three groups of countries in different level of outbreak progress, and provide informative implications for the current global pandemic.

q-bio.PE↗

arXiv2020

Hyperon Weak Radiative Decay

We revisit the hyperon weak radiative decays in the framework of the non-relativistic constituent quark model. This study confirms the nonlocal feature of the hyperon weak radiative transition operators which are dominated by the pole terms, and an overall self-consistent description of the available experimental data for the Cabibbo-favored hyperon weak radiative decays is achieved. It provides a natural mechanism for evading the Hara theorem, where sizeable parity-violating contributions can come from the intermediate orbital excitations. Cancellations between pole terms also explain significant SU(3) flavor symmetry breaking manifested by the experimental data. We also discuss several interesting selection rules arising from either the electromagnetic or the weak interaction vertices. These features suggest nontrivial relations among different hyperon decays.

hep-ph↗

arXiv2020

Hadronic weak decays of $Λ_c$ in the quark model

The hadronic weak decays of $Λ_c$ are studied in the framework of a constituent quark model. With the combined analysis of the Cabbibo-favored processes, $Λ_c\to Λπ^+$, $Σ^0π^+$ and $Σ^+π^0$, we confirm that the non-factorizable transition mechanisms play a crucial role in the understanding of their compatible branching ratios. We emphasize that the SU(3) flavor symmetry breaking effects, which is generally at the order of $1\sim 2\%$, can be amplified by the destructive interferences among the pole terms in the diagrams with internal conversion. Some contributions are sensitive to the spatial distribution of the scalar-isoscalar light-quark sector in the $Λ_c$, and its overlap with the light quarks in the final state hyperon. Namely, a compact diquark configuration is disfavored.

hep-ph↗

arXiv2020

Collaborative Fairness in Federated Learning

In current deep learning paradigms, local training or the Standalone framework tends to result in overfitting and thus poor generalizability. This problem can be addressed by Distributed or Federated Learning (FL) that leverages a parameter server to aggregate model updates from individual participants. However, most existing Distributed or FL frameworks have overlooked an important aspect of participation: collaborative fairness. In particular, all participants can receive the same or similar models, regardless of their contributions. To address this issue, we investigate the collaborative fairness in FL, and propose a novel Collaborative Fair Federated Learning (CFFL) framework which utilizes reputation to enforce participants to converge to different models, thus achieving fairness without compromising the predictive performance. Extensive experiments on benchmark datasets demonstrate that CFFL achieves high fairness, delivers comparable accuracy to the Distributed framework, and outperforms the Standalone framework.

cs.LG↗

arXiv2020

Towards Privacy-assured and Lightweight On-chain Auditing of Decentralized Storage

How to audit outsourced data in centralized storage like cloud is well-studied, but it is largely under-explored for the rising decentralized storage network (DSN) that bodes well for a billion-dollar market. To realize DSN as a usable service in a truly decentralized manner, the blockchain comes in handy -- to record and verify audit trails in forms of proof of storage, and based on that, to handle fair payments with necessary dispute resolution. Leaving the audit trails on the blockchain offers transparency and fairness, yet it 1) sacrifices privacy, as they may leak information about the data under audit, and 2) overwhelms on-chain resources, as they may be practically large in size and expensive to verify. Prior auditing designs in centralized settings are not directly applicable here. A handful of proposals targeting DSN cannot satisfactorily address these issues either. We present an auditing solution that addresses on-chain privacy and efficiency, from a synergy of homomorphic linear authenticators with polynomial commitments for succinct proofs, and the sigma protocol for provable privacy. The solution results in, per audit, 288-byte proof written to the blockchain, and constant verification cost. It can sustain long-term operation and easily scale to thousands of users on Ethereum.

cs.CR↗

arXiv2020

Crowd Counting via Segmentation Guided Attention Networks and Curriculum Loss

Automatic crowd behaviour analysis is an important task for intelligent transportation systems to enable effective flow control and dynamic route planning for varying road participants. Crowd counting is one of the keys to automatic crowd behaviour analysis. Crowd counting using deep convolutional neural networks (CNN) has achieved encouraging progress in recent years. Researchers have devoted much effort to the design of variant CNN architectures and most of them are based on the pre-trained VGG16 model. Due to the insufficient expressive capacity, the backbone network of VGG16 is usually followed by another cumbersome network specially designed for good counting performance. Although VGG models have been outperformed by Inception models in image classification tasks, the existing crowd counting networks built with Inception modules still only have a small number of layers with basic types of Inception modules. To fill in this gap, in this paper, we firstly benchmark the baseline Inception-v3 model on commonly used crowd counting datasets and achieve surprisingly good performance comparable with or better than most existing crowd counting models. Subsequently, we push the boundary of this disruptive work further by proposing a Segmentation Guided Attention Network (SGANet) with Inception-v3 as the backbone and a novel curriculum loss for crowd counting. We conduct thorough experiments to compare the performance of our SGANet with prior arts and the proposed model can achieve state-of-the-art performance with MAE of 57.6, 6.3 and 87.6 on ShanghaiTechA, ShanghaiTechB and UCF\_QNRF, respectively.

cs.CV↗

arXiv2020

Generalized Zero-Shot Domain Adaptation via Coupled Conditional Variational Autoencoders

Domain adaptation approaches aim to exploit useful information from the source domain where supervised learning examples are easier to obtain to address a learning problem in the target domain where there is no or limited availability of such examples. In classification problems, domain adaptation has been studied under varying supervised, unsupervised and semi-supervised conditions. However, a common situation when the labelled samples are available for a subset of target domain classes has been overlooked. In this paper, we formulate this particular domain adaptation problem within a generalized zero-shot learning framework by treating the labelled source domain samples as semantic representations for zero-shot learning. For this particular problem, neither conventional domain adaptation approaches nor zero-shot learning algorithms directly apply. To address this generalized zero-shot domain adaptation problem, we present a novel Coupled Conditional Variational Autoencoder (CCVAE) which can generate synthetic target domain features for unseen classes from their source domain counterparts. Extensive experiments have been conducted on three domain adaptation datasets including a bespoke X-ray security checkpoint dataset to simulate a real-world application in aviation security. The results demonstrate the effectiveness of our proposed approach both against established benchmarks and in terms of real-world applicability.

cs.CV↗

arXiv2020

Multi-Class 3D Object Detection Within Volumetric 3D Computed Tomography Baggage Security Screening Imagery

Automatic detection of prohibited objects within passenger baggage is important for aviation security. X-ray Computed Tomography (CT) based 3D imaging is widely used in airports for aviation security screening whilst prior work on automatic prohibited item detection focus primarily on 2D X-ray imagery. These works have proven the possibility of extending deep convolutional neural networks (CNN) based automatic prohibited item detection from 2D X-ray imagery to volumetric 3D CT baggage security screening imagery. However, previous work on 3D object detection in baggage security screening imagery focused on the detection of one specific type of objects (e.g., either {\it bottles} or {\it handguns}). As a result, multiple models are needed if more than one type of prohibited item is required to be detected in practice. In this paper, we consider the detection of multiple object categories of interest using one unified framework. To this end, we formulate a more challenging multi-class 3D object detection problem within 3D CT imagery and propose a viable solution (3D RetinaNet) to tackle this problem. To enhance the performance of detection we investigate a variety of strategies including data augmentation and varying backbone networks. Experimentation carried out to provide both quantitative and qualitative evaluations of the proposed approach to multi-class 3D object detection within 3D CT baggage security screening imagery. Experimental results demonstrate the combination of the 3D RetinaNet and a series of favorable strategies can achieve a mean Average Precision (mAP) of 65.3\% over five object classes (i.e. {\it bottles, handguns, binoculars, glock frames, iPods}). The overall performance is affected by the poor performance on {\it glock frames} and {\it iPods} due to the lack of data and their resemblance with the baggage clutter.

cs.CV↗

arXiv2020

Performance of quantum heat engines under the influence of long-range interactions

We examine a quantum heat engine with an interacting many-body working medium consisting of the long-range Kitaev chain to explore the role of long-range interactions in the performance of the quantum engine. By analytically studying two type of thermodynamic cycles, namely Otto cycle and Stirling cycle, we demonstrate that the work output and efficiency of long-range interacting heat engine can be boosted by long-range interactions, in comparison to the short-range counterpart. We further show that in the Otto cycle, there exist an optimal condition for which the largest enhancement of work output and efficiency can be achieved simultaneously by the long-range interactions. But, for the Stirling cycle, the condition that gives the largest enhancement in work output does not lead to the largest enhancement in efficiency. We also investigate how the parameter regimes under which the engine performance is enhanced by the long-range interactions is evolved with decrease in the range of interaction.

quant-ph↗

arXiv2020

Learning Error-Driven Curriculum for Crowd Counting

Density regression has been widely employed in crowd counting. However, the frequency imbalance of pixel values in the density map is still an obstacle to improve the performance. In this paper, we propose a novel learning strategy for learning error-driven curriculum, which uses an additional network to supervise the training of the main network. A tutoring network called TutorNet is proposed to repetitively indicate the critical errors of the main network. TutorNet generates pixel-level weights to formulate the curriculum for the main network during training, so that the main network will assign a higher weight to those hard examples than easy examples. Furthermore, we scale the density map by a factor to enlarge the distance among inter-examples, which is well known to improve the performance. Extensive experiments on two challenging benchmark datasets show that our method has achieved state-of-the-art performance.

cs.CV↗

arXiv2020

Optimizing Information Freshness via Multiuser Scheduling with Adaptive NOMA/OMA

This paper considers a wireless network with a base station (BS) conducting timely status updates to multiple clients via adaptive non-orthogonal multiple access (NOMA)/orthogonal multiple access (OMA). Specifically, the BS is able to adaptively switch between NOMA and OMA for the downlink transmission to optimize the information freshness of the network, characterized by the Age of Information (AoI) metric. If the BS chooses OMA, it can only serve one client within each time slot and should decide which client to serve; if the BS chooses NOMA, it can serve more than one client at the same time and needs to decide the power allocated to the served clients. For the simple two-client case, we formulate a Markov Decision Process (MDP) problem and develop the optimal policy for the BS to decide whether to use NOMA or OMA for each downlink transmission based on the instantaneous AoI of both clients. The optimal policy is shown to have a switching-type property with obvious decision switching boundaries. A near-optimal policy with lower computation complexity is also devised. For the more general multi-client scenario, inspired by the proposed near-optimal policy, we formulate a nonlinear optimization problem to determine the optimal power allocated to each client by maximizing the expected AoI drop of the network in each time slot. We resolve the formulated problem by approximating it as a convex optimization problem. We also derive the upper bound of the gap between the approximate convex problem and the original nonlinear, nonconvex problem. Simulation results validate the effectiveness of the adopted approximation. The performance of the adaptive NOMA/OMA scheme by solving the convex optimization is shown to be close to that of max-weight policy solved by exhaustive search...

cs.IT↗

arXiv2020

An Auto-Context Deformable Registration Network for Infant Brain MRI

Deformable image registration is fundamental to longitudinal and population analysis. Geometric alignment of the infant brain MR images is challenging, owing to rapid changes in image appearance in association with brain development. In this paper, we propose an infant-dedicated deep registration network that uses the auto-context strategy to gradually refine the deformation fields to obtain highly accurate correspondences. Instead of training multiple registration networks, our method estimates the deformation fields by invoking a single network multiple times for iterative deformation refinement. The final deformation field is obtained by the incremental composition of the deformation fields. Experimental results in comparison with state-of-the-art registration methods indicate that our method achieves higher accuracy while at the same time preserves the smoothness of the deformation fields. Our implementation is available online.

cs.CV↗

arXiv2020

Optimizing Privacy-Preserving Outsourced Convolutional Neural Network Predictions

Convolutional neural network is a machine-learning model widely applied in various prediction tasks, such as computer vision and medical image analysis. Their great predictive power requires extensive computation, which encourages model owners to host the prediction service in a cloud platform. Recent researches focus on the privacy of the query and results, but they do not provide model privacy against the model-hosting server and may leak partial information about the results. Some of them further require frequent interactions with the querier or heavy computation overheads, which discourages querier from using the prediction service. This paper proposes a new scheme for privacy-preserving neural network prediction in the outsourced setting, i.e., the server cannot learn the query, (intermediate) results, and the model. Similar to SecureML (S&P'17), a representative work that provides model privacy, we leverage two non-colluding servers with secret sharing and triplet generation to minimize the usage of heavyweight cryptography. Further, we adopt asynchronous computation to improve the throughput, and design garbled circuits for the non-polynomial activation function to keep the same accuracy as the underlying network (instead of approximating it). Our experiments on MNIST dataset show that our scheme achieves an average of 122x, 14.63x, and 36.69x reduction in latency compared to SecureML, MiniONN (CCS'17), and EzPC (EuroS&P'19), respectively. For the communication costs, our scheme outperforms SecureML by 1.09x, MiniONN by 36.69x, and EzPC by 31.32x on average. On the CIFAR dataset, our scheme achieves a lower latency by a factor of 7.14x and 3.48x compared to MiniONN and EzPC, respectively. Our scheme also provides 13.88x and 77.46x lower communication costs than MiniONN and EzPC on the CIFAR dataset.

cs.CR↗