arXiv subjects

Jie Zhang

Publications and source records attributed to Jie Zhang.

At least 829 records · Page 46Linked to original sources

arXiv2022

A Coalition Formation Game Approach for Personalized Federated Learning

Facing the challenge of statistical diversity in client local data distribution, personalized federated learning (PFL) has become a growing research hotspot. Although the state-of-the-art methods with model similarity-based pairwise collaboration have achieved promising performance, they neglect the fact that model aggregation is essentially a collaboration process within the coalition, where the complex multiwise influences take place among clients. In this paper, we first apply Shapley value (SV) from coalition game theory into the PFL scenario. To measure the multiwise collaboration among a group of clients on the personalized learning performance, SV takes their marginal contribution to the final result as a metric. We propose a novel personalized algorithm: pFedSV, which can 1. identify each client's optimal collaborator coalition and 2. perform personalized model aggregation based on SV. Extensive experiments on various datasets (MNIST, Fashion-MNIST, and CIFAR-10) are conducted with different Non-IID data settings (Pathological and Dirichlet). The results show that pFedSV can achieve superior personalized accuracy for each client, compared to the state-of-the-art benchmarks.

cs.AI↗

arXiv2022

MGC: A Complex-Valued Graph Convolutional Network for Directed Graphs

Recent advancements in Graph Neural Networks have led to state-of-the-art performance on graph representation learning. However, the majority of existing works process directed graphs by symmetrization, which causes loss of directional information. To address this issue, we introduce the magnetic Laplacian, a discrete Schrödinger operator with magnetic field, which preserves edge directionality by encoding it into a complex phase with an electric charge parameter. By adopting a truncated variant of PageRank named Linear- Rank, we design and build a low-pass filter for homogeneous graphs and a high-pass filter for heterogeneous graphs. In this work, we propose a complex-valued graph convolutional network named Magnetic Graph Convolutional network (MGC). With the corresponding complex-valued techniques, we ensure our model will be degenerated into real-valued when the charge parameter is in specific values. We test our model on several graph datasets including directed homogeneous and heterogeneous graphs. The experimental results demonstrate that MGC is fast, powerful, and widely applicable.

cs.LG↗

arXiv2022

Experimental observations of marginal criticality in granular materials

Two drastically different theories predict the marginal criticality of jamming. The full replica symmetry breaking (fullRSB) theory [1-4] predicts the power-law distributions of weak contact forces and small inter-particle gaps in infinite-dimensional hard-sphere glass, with two nontrivial exponents $θ_f=0.42311...$ and $γ=0.41269...$, respectively. While the marginal mechanical stability (MMS) analysis [5-8] predicts that the isostatic random packings of hard frictionless spheres under external stress are marginally stable and provides inequality relationships for the exponents of the weak-force and inter-particle-gap distributions. Here we measure precisely contact forces and particle positions in isotropic jammed bidisperse photoelastic disks and find the clear power-law distributions of weak forces and small inter-particle gaps, with both exponents $θ_f=0.44(2)$ and $γ= 0.43(3) $ in an excellent agreement with the fullRSB theory. As the jammed packing subject to area-conserved cyclic pure shear approaches the yielding point, the two exponents change substantially from those of the isotropic case but they still satisfy the scaling relationship provided by the MMS argument. Our results provide strong experimental evidences for the robustness of the infinite-dimensional theory and the MMS analysis in real-world amorphous materials.

cond-mat.soft↗

arXiv2022

Three-dimensional dynamics of a pair of deformable bubbles rising initially in line. Part 2: Highly inertial regimes

The buoyancy-driven dynamics of a pair of gas bubbles released in line is investigated numerically, focusing on highly inertial conditions under which isolated bubbles follow non-straight paths. In an early stage, the second bubble always drifts out of the wake of the leading one. Then, depending on the ratios of the buoyancy, viscous and capillary forces which define the Galilei ($Ga$) and Bond ($Bo$) numbers of the system, five distinct regimes specific to such conditions are identified, in which the two bubbles may rise independently or continue to interact and possibly collide in the end. In the former case, they usually perform large-amplitude planar zigzags within the same plane or within two distinct planes, depending on the oblateness of the leading bubble. However, for large enough $Ga$ and low enough $Bo$, they follow nearly vertical paths with small-amplitude erratic horizontal deviations. Increasing $Bo$ makes the wake-induced attraction toward the leading bubble stronger, forcing the two bubbles to realign vertically one or more times along their ascent. During such sequences, wake vortices may hit the trailing bubble, deflecting its path and, depending on the case, promoting or hindering further possibilities of interaction. In some regimes, varying the initial distance separating the two bubbles modifies their lateral separation beyond the initial stage. Similarly, minute initial angular deviations favour the selection of a single vertical plane of rise common to both bubbles. These changes may dramatically affect the fate of the tandem as, depending on $Bo$, they promote or prevent future vertical realignments.

physics.flu-dyn↗

arXiv2022

Supervised and Self-supervised Pretraining Based COVID-19 Detection Using Acoustic Breathing/Cough/Speech Signals

In this work, we propose a bi-directional long short-term memory (BiLSTM) network based COVID-19 detection method using breath/speech/cough signals. By using the acoustic signals to train the network, respectively, we can build individual models for three tasks, whose parameters are averaged to obtain an average model, which is then used as the initialization for the BiLSTM model training of each task. This initialization method can significantly improve the performance on the three tasks, which surpasses the official baseline results. Besides, we also utilize a public pre-trained model wav2vec2.0 and pre-train it using the official DiCOVA datasets. This wav2vec2.0 model is utilized to extract high-level features of the sound as the model input to replace conventional mel-frequency cepstral coefficients (MFCC) features. Experimental results reveal that using high-level features together with MFCC features can improve the performance. To further improve the performance, we also deploy some preprocessing techniques like silent segment removal, amplitude normalization and time-frequency mask. The proposed detection model is evaluated on the DiCOVA dataset and results show that our method achieves an area under curve (AUC) score of 88.44% on blind test in the fusion track.

eess.AS↗

arXiv2022

A Noise-Robust Self-supervised Pre-training Model Based Speech Representation Learning for Automatic Speech Recognition

Wav2vec2.0 is a popular self-supervised pre-training framework for learning speech representations in the context of automatic speech recognition (ASR). It was shown that wav2vec2.0 has a good robustness against the domain shift, while the noise robustness is still unclear. In this work, we therefore first analyze the noise robustness of wav2vec2.0 via experiments. We observe that wav2vec2.0 pre-trained on noisy data can obtain good representations and thus improve the ASR performance on the noisy test set, which however brings a performance degradation on the clean test set. To avoid this issue, in this work we propose an enhanced wav2vec2.0 model. Specifically, the noisy speech and the corresponding clean version are fed into the same feature encoder, where the clean speech provides training targets for the model. Experimental results reveal that the proposed method can not only improve the ASR performance on the noisy test set which surpasses the original wav2vec2.0, but also ensure a tiny performance decrease on the clean test set. In addition, the effectiveness of the proposed method is demonstrated under different types of noise conditions.

eess.AS↗

arXiv2022

GraphCode2Vec: Generic Code Embedding via Lexical and Program Dependence Analyses

Code embedding is a keystone in the application of machine learning on several Software Engineering (SE) tasks. To effectively support a plethora of SE tasks, the embedding needs to capture program syntax and semantics in a way that is generic. To this end, we propose the first self-supervised pre-training approach (called GraphCode2Vec) which produces task-agnostic embedding of lexical and program dependence features. GraphCode2Vec achieves this via a synergistic combination of code analysis and Graph Neural Networks. GraphCode2Vec is generic, it allows pre-training, and it is applicable to several SE downstream tasks. We evaluate the effectiveness of GraphCode2Vec on four (4) tasks (method name prediction, solution classification, mutation testing and overfitted patch classification), and compare it with four (4) similarly generic code embedding baselines (Code2Seq, Code2Vec, CodeBERT, GraphCodeBERT) and 7 task-specific, learning-based methods. In particular, GraphCode2Vec is more effective than both generic and task-specific learning-based baselines. It is also complementary and comparable to GraphCodeBERT (a larger and more complex model). We also demonstrate through a probing and ablation study that GraphCode2Vec learns lexical and program dependence features and that self-supervised pre-training improves effectiveness.

cs.SE↗

arXiv2022

The Preliminary design of DC Magnet Power Supply System for ITER Static Magnetic Field Test facility

ITER (International Thermonuclear Experimental Reactor) static magnetic field (SMF) test facility requires a DC power supply with low voltage, high current, and high stability. Due to the limitation ofswitching loss, there is a contradiction between the output current capability and the output ripple. Large output current usually leads to low switching frequency, and low switching frequency will generate a large number of harmonics. To solve the problems, a topology based on the interleaving parallel buck converter is used and tested in this paper. Moreover, the topology is realized with only a small number of switching metal-oxide-semiconductor field effect transistors (MOSFETs). This article introduces the system design scheme and control method in detail. The analysis of harmonic and simulation are carried out. The validity of proposed scheme and control strategy were confirmed by experiments, the power supply system can supply large current of 15kA and has ability of low ripple.

physics.ins-det↗

arXiv2022

High-Uniformity Calculation Method of Four-Coil Configuration in Large-Caliber Magnetic Field Immunity Testing System

Power electronic equipment regulated by the International Thermonuclear Experimental Reactor (ITER) organization must pass the relevant steady-state magnetic field immunity test. The main body of magnetic field immunity test is magnetic field generator coil. Through mathematical derivation in this paper, the magnetic field calculation formulas of four-coil configuration under ideal and actual models are obtained. The traditional method of magnetic field performance calculation is compared with the general formula method under the ideal model. A global parameter optimization method based on Lagrange Multiplier by KKT conditions is proposed to obtain the coil parameters of high-uniformity magnetic field. The magnetic field distribution in the uniform zone is revealed by the finite element method. The model analysis is proved to be correct and effective by experimental results. The research of this paper provides a practical scheme for the coil design with high magnetic field and high-quality uniformity.

physics.ins-det↗

arXiv2022

Semi-Dirac and Weyl Fermions in Transition Metal Oxides

We show that a class of compounds with $I$4/$mcm$ crystalline symmetry hosts three-dimensional semi-Dirac fermions. Unlike the known two-dimensional semi-Dirac points, the degeneracy of these three-dimensional semi-Dirac points is not lifted by spin-orbit coupling due to the protection by a nonsymmorphic symmetry -- screw rotation in the $a-b$ plane and a translation along the $c$ axis. This crystalline symmetry is found in tetragonal perovskite oxides, realizable in thin films by epitaxial strain that results in a$^0$a$^0$c$^-$-type octahedral rotation. Interestingly, with broken time-reversal symmetry, two pairs of Weyl points emerge from the semi-Dirac points within the Brillouin zone, and an additional lattice distortion leads to enhanced intrinsic anomalous Hall effect. The ability to tune the Berry phase by epitaxial strain can be useful in novel oxide-based electronic devices.

cond-mat.str-el↗

arXiv2021

Learning Variable Ordering Heuristics for Solving Constraint Satisfaction Problems

Backtracking search algorithms are often used to solve the Constraint Satisfaction Problem (CSP). The efficiency of backtracking search depends greatly on the variable ordering heuristics. Currently, the most commonly used heuristics are hand-crafted based on expert knowledge. In this paper, we propose a deep reinforcement learning based approach to automatically discover new variable ordering heuristics that are better adapted for a given class of CSP instances. We show that directly optimizing the search cost is hard for bootstrapping, and propose to optimize the expected cost of reaching a leaf node in the search tree. To capture the complex relations among the variables and constraints, we design a representation scheme based on Graph Neural Network that can process CSP instances with different sizes and constraint arities. Experimental results on random CSP instances show that the learned policies outperform classical hand-crafted heuristics in terms of minimizing the search tree size, and can effectively generalize to instances that are larger than those used in training.

cs.AI↗

arXiv2021

Initiative Defense against Facial Manipulation

Benefiting from the development of generative adversarial networks (GAN), facial manipulation has achieved significant progress in both academia and industry recently. It inspires an increasing number of entertainment applications but also incurs severe threats to individual privacy and even political security meanwhile. To mitigate such risks, many countermeasures have been proposed. However, the great majority methods are designed in a passive manner, which is to detect whether the facial images or videos are tampered after their wide propagation. These detection-based methods have a fatal limitation, that is, they only work for ex-post forensics but can not prevent the engendering of malicious behavior. To address the limitation, in this paper, we propose a novel framework of initiative defense to degrade the performance of facial manipulation models controlled by malicious users. The basic idea is to actively inject imperceptible venom into target facial data before manipulation. To this end, we first imitate the target manipulation model with a surrogate model, and then devise a poison perturbation generator to obtain the desired venom. An alternating training strategy are further leveraged to train both the surrogate model and the perturbation generator. Two typical facial manipulation tasks: face attribute editing and face reenactment, are considered in our initiative defense framework. Extensive experiments demonstrate the effectiveness and robustness of our framework in different settings. Finally, we hope this work can shed some light on initiative countermeasures against more adversarial scenarios.

cs.CV↗

arXiv2021

GIMIRec: Global Interaction Information Aware Multi-Interest Framework for Sequential Recommendation

Sequential recommendation based on multi-interest framework models the user's recent interaction sequence into multiple different interest vectors, since a single low-dimensional vector cannot fully represent the diversity of user interests. However, most existing models only intercept users' recent interaction behaviors as training data, discarding a large amount of historical interaction sequences. This may raise two issues. On the one hand, data reflecting multiple interests of users is missing; on the other hand, the co-occurrence between items in historical user-item interactions is not fully explored. To tackle the two issues, this paper proposes a novel sequential recommendation model called "Global Interaction Aware Multi-Interest Framework for Sequential Recommendation (GIMIRec)". Specifically, a global context extraction module is firstly proposed without introducing any external information, which calculates a weighted co-occurrence matrix based on the constrained co-occurrence number of each item pair and their time interval from the historical interaction sequences of all users and then obtains the global context embedding of each item by using a simplified graph convolution. Secondly, the time interval of each item pair in the recent interaction sequence of each user is captured and combined with the global context item embedding to get the personalized item embedding. Finally, a self-attention based multi-interest framework is applied to learn the diverse interests of users for sequential recommendation. Extensive experiments on the three real-world datasets of Amazon-Books, Taobao-Buy and Amazon-Hybrid show that the performance of GIMIRec on the Recall, NDCG and Hit Rate indicators is significantly superior to that of the state-of-the-art methods. Moreover, the proposed global context extraction module can be easily transplanted to most sequential recommendation models.

cs.IR↗

arXiv2021

Tracing Text Provenance via Context-Aware Lexical Substitution

Text content created by humans or language models is often stolen or misused by adversaries. Tracing text provenance can help claim the ownership of text content or identify the malicious users who distribute misleading content like machine-generated fake news. There have been some attempts to achieve this, mainly based on watermarking techniques. Specifically, traditional text watermarking methods embed watermarks by slightly altering text format like line spacing and font, which, however, are fragile to cross-media transmissions like OCR. Considering this, natural language watermarking methods represent watermarks by replacing words in original sentences with synonyms from handcrafted lexical resources (e.g., WordNet), but they do not consider the substitution's impact on the overall sentence's meaning. Recently, a transformer-based network was proposed to embed watermarks by modifying the unobtrusive words (e.g., function words), which also impair the sentence's logical and semantic coherence. Besides, one well-trained network fails on other different types of text content. To address the limitations mentioned above, we propose a natural language watermarking scheme based on context-aware lexical substitution (LS). Specifically, we employ BERT to suggest LS candidates by inferring the semantic relatedness between the candidates and the original sentence. Based on this, a selection strategy in terms of synchronicity and substitutability is further designed to test whether a word is exactly suitable for carrying the watermark signal. Extensive experiments demonstrate that, under both objective and subjective metrics, our watermarking scheme can well preserve the semantic integrity of original sentences and has a better transferability than existing methods. Besides, the proposed LS approach outperforms the state-of-the-art approach on the Stanford Word Substitution Benchmark.

cs.CR↗

arXiv2021

Resolving Two Distinct Thermal X-ray Components in A compound Solar Flare

X-ray emission provides the most direct diagnostics of the energy-release process in solar flares. Occasionally, a superhot X-ray source is found to be above hot flare loops of ~10 MK temperature. While the origin of the superhot plasma is still elusive, it has conjured up an intriguing image of in-situ plasma heating near the reconnection site high above the flare loops, in contrast to the conventional picture of chromospheric evaporation. Here we investigate an extremely long-duration solar flare, in which EUV images show two distinct flare loop systems that appear successively along a Gamma-shaped polarity inversion line (PIL). When both flare loop systems are present, the HXR spectrum is found to be well fitted by combining a hot component (Te ~12 MK) and a superhot component (Te ~30 MK). Associated with a fast CME, the superhot X-ray source is located at top of the flare arcade that appears earlier, straddling and extending along the long "arm" of the Gamma-shaped PIL. Associated with a slow CME, the hot X-ray source is located at the top of the flare arcade that appears later and sits astride the short "arm" of the Gamma-shaped PIL. Aided by observations from a different viewing angle, we are able to verify that the superhot X-ray source is above the hot one in projection, but the two sources belong to different flare loop systems. Thus, this case study provides a stereoscopic observation explaining the co-existence of superhot and hot X-ray emitting plasmas in solar flares.

astro-ph.SR↗

arXiv2021

Peer-to-Peer Energy Trading in a Microgrid Leveraged by Smart Contracts

The current electricity networks were not initially designed for the high integration of variable generation technologies. They suffer significant losses due to the combustion of fossil fuels, the long-distance transmission, and distribution of the power to the network. Recently, \emph{prosumers}, both consumers and producers, emerge with the increasing affordability to invest in domestic solar systems. Prosumers may trade within their communities to better manage their demand and supply as well as providing social and economic benefits. In this paper, we explore the use of Blockchain technologies and auction mechanisms to facilitate autonomous peer-to-peer energy trading within microgrids. We design two frameworks that utilize the smart contract functionality in Ethereum and employ the continuous double auction and uniform-price double-sided auction mechanisms, respectively. We validate our design by conducting A/B tests to compare the performance of different frameworks on a real-world dataset. The key characteristics of the two frameworks and several cost analyses are presented for comparison. Our results demonstrate that a P2P trading platform that integrates the blockchain technologies and agent-based systems is promising to complement the current centralized energy grid. We also identify a number of limitations, alternative solutions, and directions for future work.

cs.MA↗

arXiv2021

Parallel-Electromagnetically-Induced-Transparency Near Ground-State Cooling of a Trapped-ion Crystal

We theoretically propose and experimentally demonstrate a parallel-electromagnetically-induced transparency (parallel-EIT) cooling technique for ion crystals in the Paul trap. It has less stringent requirements on the cooling resonance condition than the standard electromagnetically-induced transparency (EIT) cooling, thus allowing, in principle, to simultaneously cool the motional mode spectrum with an arbitrary range. A proof-of-principle validation for this cooling scheme is experimentally demonstrated with up to 4 trapped 40Ca+ ions. We observe simultaneous near-ground-state cooling for all motional modes with best average phonon number about 0.2. By tuning the trap frequency in a large range to imitate a broadband motional mode spectrum, we can still reach almost the same cooling limit for all the modes while standard EIT cooling shows limited cooling range. Our method has a simple experimental configuration, requiring only appropriate modulation of the probe beam of standard EIT cooling, and can be applied to various types of ions (e.g., 171Yb+, 40Ca+). This cooling scheme provides a powerful tool for the initialization of the trapped-ion quantum computers and simulators.

quant-ph↗

arXiv2021

The Smoothing Proximal Gradient Algorithm with Extrapolation for the Relaxation of $l_0$ Regularization Problem

In this paper, we consider the exact continuous relaxation model of $l_0$ regularization problem which was given by Bian and Chen (SIAM J. Numer. Anal 58(1): 858-883, 2020) and propose a smoothing proximal gradient algorithm with extrapolation (SPGE) for this kind of problem. We show that any accumulation point of the sequence generated by SPGE algorithm is a lifted stationary point of the relaxation model for a fairly general choice of extrapolation parameter. Moreover, it is shown that the whole sequence generated by SPGE algorithm converges to a lifted stationary point of the relaxation model. The convergence rate of $o(\frac{1}{k^{1-σ}})(σ\in(0,1))$ with respect to squared proximal residual is established. Finally, we conduct three important numerical experiments to illustrate the faster efficiency of the SPGE algorithm compared with the smoothing proximal gradient(SPG) algorithm proposed by Bian and Chen (SIAM J. Numer. Anal 58(1):858-883, 2020).

math.OC↗