arXiv subjects

Jie Zhang

Publications and source records attributed to Jie Zhang.

At least 775 records · Page 43Linked to original sources

arXiv2022

Speech Enhancement Using Self-Supervised Pre-Trained Model and Vector Quantization

With the development of deep learning, neural network-based speech enhancement (SE) models have shown excellent performance. Meanwhile, it was shown that the development of self-supervised pre-trained models can be applied to various downstream tasks. In this paper, we will consider the application of the pre-trained model to the real-time SE problem. Specifically, the encoder and bottleneck layer of the DEMUCS model are initialized using the self-supervised pretrained WavLM model, the convolution in the encoder is replaced by causal convolution, and the transformer encoder in the bottleneck layer is based on causal attention mask. In addition, as discretizing the noisy speech representations is more beneficial for denoising, we utilize a quantization module to discretize the representation output from the bottleneck layer, which is then fed into the decoder to reconstruct the clean speech waveform. Experimental results on the Valentini dataset and an internal dataset show that the pre-trained model based initialization can improve the SE performance and the discretization operation suppresses the noise component in the representations to some extent, which can further improve the performance.

eess.AS↗

arXiv2022

Federated Learning with Label Distribution Skew via Logits Calibration

Traditional federated optimization methods perform poorly with heterogeneous data (ie, accuracy reduction), especially for highly skewed data. In this paper, we investigate the label distribution skew in FL, where the distribution of labels varies across clients. First, we investigate the label distribution skew from a statistical view. We demonstrate both theoretically and empirically that previous methods based on softmax cross-entropy are not suitable, which can result in local models heavily overfitting to minority classes and missing classes. Additionally, we theoretically introduce a deviation bound to measure the deviation of the gradient after local update. At last, we propose FedLC (\textbf {Fed} erated learning via\textbf {L} ogits\textbf {C} alibration), which calibrates the logits before softmax cross-entropy according to the probability of occurrence of each class. FedLC applies a fine-grained calibrated cross-entropy loss to local update by adding a pairwise label margin. Extensive experiments on federated datasets and real-world datasets demonstrate that FedLC leads to a more accurate global model and much improved performance. Furthermore, integrating other FL methods into our approach can further enhance the performance of the global model.

cs.LG↗

arXiv2022

Experimental violation of Leggett-Garg inequality in a three-level trapped-ion system

Leggett-Garg inequality (LGI) studies the temporal correlation in the evolution of physical systems. Classical systems obey the LGI but quantum systems may violate it. The extent of the violation depends on the dimension of the quantum system and the state update rule. In this work, we experimentally test the LGI in a three-level trapped-ion system under the model of a large spin precessing in a magnetic field. The Von Neumann and Lüders state update rules are employed in our system for direct comparative analysis. The maximum observed value of Leggett-Garg correlator under the Von Neumann state update rule is $K_3 = 1.739 \pm 0.014$, which demonstrates a violation of the Lüders bound by 17 standard deviations and is by far the most significant violation in natural three-level systems.

quant-ph↗

arXiv2022

Learning to Solve Multiple-TSP with Time Window and Rejections via Deep Reinforcement Learning

We propose a manager-worker framework based on deep reinforcement learning to tackle a hard yet nontrivial variant of Travelling Salesman Problem (TSP), \ie~multiple-vehicle TSP with time window and rejections (mTSPTWR), where customers who cannot be served before the deadline are subject to rejections. Particularly, in the proposed framework, a manager agent learns to divide mTSPTWR into sub-routing tasks by assigning customers to each vehicle via a Graph Isomorphism Network (GIN) based policy network. A worker agent learns to solve sub-routing tasks by minimizing the cost in terms of both tour length and rejection rate for each vehicle, the maximum of which is then fed back to the manager agent to learn better assignments. Experimental results demonstrate that the proposed framework outperforms strong baselines in terms of higher solution quality and shorter computation time. More importantly, the trained agents also achieve competitive performance for solving unseen larger instances.

cs.LG↗

arXiv2022

Class-Incremental Learning via Knowledge Amalgamation

Catastrophic forgetting has been a significant problem hindering the deployment of deep learning algorithms in the continual learning setting. Numerous methods have been proposed to address the catastrophic forgetting problem where an agent loses its generalization power of old tasks while learning new tasks. We put forward an alternative strategy to handle the catastrophic forgetting with knowledge amalgamation (CFA), which learns a student network from multiple heterogeneous teacher models specializing in previous tasks and can be applied to current offline methods. The knowledge amalgamation process is carried out in a single-head manner with only a selected number of memorized samples and no annotations. The teachers and students do not need to share the same network structure, allowing heterogeneous tasks to be adapted to a compact or sparse data representation. We compare our method with competitive baselines from different strategies, demonstrating our approach's advantages.

cs.LG↗

arXiv2022

A Multi-Channel Next POI Recommendation Framework with Multi-Granularity Check-in Signals

Current study on next POI recommendation mainly explores user sequential transitions with the fine-grained individual-user POI check-in trajectories only, which suffers from the severe check-in data sparsity issue. In fact, coarse-grained signals (i.e., region- and global-level check-ins) in such sparse check-ins would also benefit to augment user preference learning. Specifically, our data analysis unveils that user movement exhibits noticeable patterns w.r.t. the regions of visited POIs. Meanwhile, the global all-user check-ins can help reflect sequential regularities shared by the crowd. We are, therefore, inspired to propose the MCMG: a Multi-Channel next POI recommendation framework with Multi-Granularity signals categorized from two orthogonal perspectives, i.e., fine-coarse grained check-ins at either POI/region level or local/global level. Being equipped with three modules (i.e., global user behavior encoder, local multi-channel encoder, and region-aware weighting strategy), MCMG is capable of capturing both fine- and coarse-grained sequential regularities as well as exploring the dynamic impact of multi-channel by differentiating the region check-in patterns. Extensive experiments on four real-world datasets show that our MCMG significantly outperforms state-of-the-art next POI recommendation approaches.

cs.IR↗

arXiv2022

Adversarial Examples for Good: Adversarial Examples Guided Imbalanced Learning

Adversarial examples are inputs for machine learning models that have been designed by attackers to cause the model to make mistakes. In this paper, we demonstrate that adversarial examples can also be utilized for good to improve the performance of imbalanced learning. We provide a new perspective on how to deal with imbalanced data: adjust the biased decision boundary by training with Guiding Adversarial Examples (GAEs). Our method can effectively increase the accuracy of minority classes while sacrificing little accuracy on majority classes. We empirically show, on several benchmark datasets, our proposed method is comparable to the state-of-the-art method. To our best knowledge, we are the first to deal with imbalanced learning with adversarial examples.

cs.LG↗

arXiv2022

Modeling of bound electron effects in particle-in-cell simulation

To include the bound electron effects in particle-in-cell (PIC) simulation, we propose a model in which the response of the dipole components of partially ionized ions to external electromagnetic fields can be included. Instead of treating the macro-ion particle as a single particle without an internal structure, the ions are considered to have a structure composed of a central nucleus and a bounded electron cloud in our model. The two parts experience the interactions of both the external electromagnetic fields and the internal Coulomb fields. In this way, the laser scattering effects by a partially ionized medium can be modeled properly in the PIC simulation. The laser propagation in a neutral medium and the Bragg scattering of the laser in crystal structure have been simulated with a PIC code modified based on our model as the benchmark. Our model may find applications to study some interesting problems, such as the x-ray laser-driven wakefield acceleration in crystals, the x-ray laser-driven high energy density physics, and intense laser propagation in partially ionized nonlinear optical materials, etc.

physics.plasm-ph↗

arXiv2022

Multi-modal Transformer Path Prediction for Autonomous Vehicle

Reasoning about vehicle path prediction is an essential and challenging problem for the safe operation of autonomous driving systems. There exist many research works for path prediction. However, most of them do not use lane information and are not based on the Transformer architecture. By utilizing different types of data collected from sensors equipped on the self-driving vehicles, we propose a path prediction system named Multi-modal Transformer Path Prediction (MTPP) that aims to predict long-term future trajectory of target agents. To achieve more accurate path prediction, the Transformer architecture is adopted in our model. To better utilize the lane information, the lanes which are in opposite direction to target agent are not likely to be taken by the target agent and are consequently filtered out. In addition, consecutive lane chunks are combined to ensure the lane input to be long enough for path prediction. An extensive evaluation is conducted to show the efficacy of the proposed system using nuScene, a real-world trajectory forecasting dataset.

cs.RO↗

arXiv2022

Poison Ink: Robust and Invisible Backdoor Attack

Recent research shows deep neural networks are vulnerable to different types of attacks, such as adversarial attack, data poisoning attack and backdoor attack. Among them, backdoor attack is the most cunning one and can occur in almost every stage of deep learning pipeline. Therefore, backdoor attack has attracted lots of interests from both academia and industry. However, most existing backdoor attack methods are either visible or fragile to some effortless pre-processing such as common data transformations. To address these limitations, we propose a robust and invisible backdoor attack called "Poison Ink". Concretely, we first leverage the image structures as target poisoning areas, and fill them with poison ink (information) to generate the trigger pattern. As the image structure can keep its semantic meaning during the data transformation, such trigger pattern is inherently robust to data transformations. Then we leverage a deep injection network to embed such trigger pattern into the cover image to achieve stealthiness. Compared to existing popular backdoor attack methods, Poison Ink outperforms both in stealthiness and robustness. Through extensive experiments, we demonstrate Poison Ink is not only general to different datasets and network architectures, but also flexible for different attack scenarios. Besides, it also has very strong resistance against many state-of-the-art defense techniques.

cs.CR↗

arXiv2022

Microwave QR Code: An IRS-Based Solution

This letter proposes to employ intelligent reflecting surface (IRS) as an information media to display a microwave quick response (QR) code for Internet-of-Things applications. To be specific, an IRS is used to form a dynamic bitmap image thanks to its tunable elements. With a QR code shown on the IRS, the transmitting and receiving antenna arrays are jointly designed to scan it by radiating electromagnetic wave as well as receiving and detecting the reflected signal. Based on such an idea, an IRS enabled information and communication system is modelled. Accordingly, some fundamental systematic operating mechanisms are investigated, involving derivation of average bit error probability for signal modulation, QR code implementation on an IRS, transmission design, detection, etc. The simulations are performed to show the achievable communication performance of system and confirm the feasibility of IRS-based microwave QR code.

eess.SP↗

arXiv2022

Tail-wave-assisted Positron Acceleration in Nonlinear Laser Plasma Wakefields

Relativistic laser wakefield acceleration is characterized by an unsurpassed accelerating gradient, which is very suitable for electron acceleration over short distances and could be a promising candidate for next-generation compact accelerators. However, using this technique for positron acceleration is still challenging because positively charged particles are naturally defocused in nonlinear wakefields. Here we propose and numerically demonstrate a scheme to accelerate an externally injected positron beam in a nonlinear laser wakefield in a regime where a tail wave is formed behind density cusps of the wakefield. This tail wave can provide a focusing force in addition to longitudinal acceleration for the positrons. Three-dimensional particle-in-cell simulations demonstrate that a trapping efficiency of positrons of nearly 100% in the nonlinear wakefield is possible. This scheme may open a simple way for compact positron acceleration to multi-100 MeV with terawatt-class laser systems at high repetition rates without the need for special laser modes and plasma structures.

physics.plasm-ph↗

arXiv2022

Experimental and Numerical Verification of Anomalous Screening Theory in Granular Matter

The concept of mechanical screening is widely applied in solid-state systems. Examples include nucleation of defects in crystalline materials, scars and pleats in curved crystals, wrinkles in strongly confined thin sheets, and cell-rearrangements in biological tissue. Available theories of such screening usually contain a crucial ingredient, which is the existence of an ordered reference state, with respect to which screening elements nucleate to release stresses. In contradistinction, amorphous materials in which a unique reference state does not exist, nevertheless exhibits plastic events that act as screening geometric charges with significant implications on the mechanical response. In a recent paper [Phys. Rev. E 104, 024904] it was proposed that mechanical strains in amorphous solids can be either weakly or strongly screened by the formation of low or high density of plastic events. At low densities the screening effect is reminiscent of the role of dipoles in dielectrics, in only renormalizing the elastic moduli. The effect of high density screening has no immediate electrostatic analog and is expected to change qualitatively the mechanical response, as seen for example in the displacement field. On the basis of experiments and simulations, we show that in granular matter, strong screening results in significant deviation from elasticity theory. The theoretical analysis, which accounts for an emergent inherent length scale, the experimental measurements and the numerical simulations of frictional granular amorphous assemblies are in agreement with each other, and provide a strong support for the novel continuum theory.

cond-mat.dis-nn↗

arXiv2022

An IRS Backscatter Enabled Integrated Sensing, Communication and Computation System

This paper proposes to leverage intelligent reflecting surface (IRS) backscatter to realize radio-frequency-chain-free uplink-transmissions (RFCF-UT). In this communication paradigm, IRS works as an information carrier, whose elements are capable of adjusting their amplitudes and phases to collaboratively portray an electromagnetic image like a dynamic quick response (QR) code, rather than a familiar reflection device, while a full-duplex base station (BS) is used as a scanner to collect and recognize the information on IRS. To elaborate it, an integrated sensing, communication and computation system as an example is presented, in which a dual-functional radar-communication BS simultaneously detects the target and collects the data from user equipments each connected to an IRS. Based on the established model, partial and binary data offloading strategies are respectively considered. By defining a performance metric named weighted throughput capacity (WTC), two maximization problems of WTC are formulated. According to the coupling degree of optimization variables in the objective function and the constraints, each optimization problem is firstly decomposed into two subproblems. Then, the methods of linear programming, fractional programming, integer programming and alternative optimization are developed to solve the subproblems. The simulation results demonstrate the achievable WTC of the considered system, thereby validating RFCF-UT.

eess.SP↗

arXiv2022

Direct numerical simulations of incompressible multiphase electrohydrodynamic flow with single-phase transportation schemes

In the present study, two schemes named face discernment and flux correction are proposed to achieve single-phase transportation of free charge in multiphase electrohydrodynamic(EHD) problems. Many EHD phenomena occur between air and another liquid while the free charge can only be transported in the liquid phase through ohmic conduction and convection due to the poor conductivity of air. However, the charge may be leaked into the dielectric air during the simulation due to the asynchronous transportation between interface and free charge. To avoid this unphysical error, a face discernment method is designed to produce an accurate ohmic conduction of free charge by providing a superior physical properties distribution at the interface. Subsequently the flux correction method is developed to correct the advection flux of charge density to prevent ions crossing the interface. These two schemes are based on the Volume of Fluid (VOF) model and independent with the specific interface updating method. The performance of the proposed methods are carefully validated with several test cases. The algorithms are implemented as an OpenFOAM extension and are published as open source.

physics.flu-dyn↗

arXiv2022

Bunched proton acceleration from a laser-irradiated cone target

Laser-driven ion acceleration is an attractive technique for compact high-energy ion sources. Currently, among various physical and technical issues to be solved, the boost of ion energy and the reduction of energy spread represent the key challenges with this technique. Here we present a scheme to tackle these challenges by using a hundred-terawatt-class laser pulse irradiating a cone target. Three-dimensional particle-in-cell simulations show that a large number of electrons are dragged out of the cone walls and accelerated to hundreds of MeV by the laser fields inside the cone. When these energetic dense electron beams pass through the cone target tip into vacuum, a very high bunching acceleration field, up to tens of TV/m, quickly forms. Protons are accelerated and simultaneously bunched by this field, resulting in quasi-monoenergetic proton beams with hundred MeV energy and low energy spread of ~2%. Results exploring the scaling of the proton beam energy with laser and target parameters are presented, indicating that the scheme is robust. This opens a new route for compact high-energy proton sources from fundamental research to biomedical applications.

physics.plasm-ph↗

arXiv2022

Radio detection of an elusive millisecond pulsar in the Globular Cluster NGC 6397

We report the discovery of a new 5.78 ms-period millisecond pulsar (MSP), PSR J1740-5340B (NGC 6397B), in an eclipsing binary system discovered with the Parkes radio telescope (now also known as Murriyang), Australia, and confirmed with the MeerKAT radio telescope in South Africa. The measured orbital period, 1.97 days, is the longest among all eclipsing binaries in globular clusters (GCs) and consistent with that of the coincident X-ray source U18, previously suggested to be a 'hidden MSP'. Our XMM-Newton observations during NGC 6397B's radio quiescent epochs detected no X-ray flares. NGC 6397B is either a transitional MSP or an eclipsing binary in its initial stage of mass transfer after the companion star left the main sequence. The discovery of NGC 6397B potentially reveals a subgroup of extremely faint and heavily obscured binary pulsars, thus providing a plausible explanation to the apparent dearth of binary neutron stars in core-collapsed GCs as well as a critical constraint on the evolution of GCs.

astro-ph.HE↗

arXiv2022

Learning Contextually Fused Audio-visual Representations for Audio-visual Speech Recognition

With the advance in self-supervised learning for audio and visual modalities, it has become possible to learn a robust audio-visual speech representation. This would be beneficial for improving the audio-visual speech recognition (AVSR) performance, as the multi-modal inputs contain more fruitful information in principle. In this paper, based on existing self-supervised representation learning methods for audio modality, we therefore propose an audio-visual representation learning approach. The proposed approach explores both the complementarity of audio-visual modalities and long-term context dependency using a transformer-based fusion module and a flexible masking strategy. After pre-training, the model is able to extract fused representations required by AVSR. Without loss of generality, it can be applied to single-modal tasks, e.g. audio/visual speech recognition by simply masking out one modality in the fusion module. The proposed pre-trained model is evaluated on speech recognition and lipreading tasks using one or two modalities, where the superiority is revealed.

eess.IV↗