Searcharxiv⌕ Search

arXiv subjects

Peng Ye

Publications and source records attributed to Peng Ye.

At least 163 records · Page 9Linked to original sources
arXiv2023

Real-time frequency measurement based on parallel pipeline FFT for time-stretched acquisition system

Real-time frequency measurement for non-repetitive and statistically rare signals are challenging problems in the electronic measurement area, which places high demands on the bandwidth, sampling rate, data processing and transmission capabilities of the measurement system. The time-stretching sampling system overcomes the bandwidth limitation and sampling rate limitation of electronic digitizers, allowing continuous ultra-high-speed acquisition at refresh rates of billions of frames per second. However, processing the high sampling rate signals of hundreds of GHz is an extremely challenging task, which becomes the bottleneck of the real-time analysis for non-stationary signals. In this work, a real-time frequency measurement system is designed based on a parallel pipelined FFT structure. Tens of FFT channels are pipelined to process the incoming high sampling rate signals in sequence, and a simplified parabola fitting algorithm is implemented in the FFT channel to improve the frequency precision. The frequency results of these FFT channels are reorganized and finally uploaded to an industrial personal computer for visualization and offline data mining. A real-time transmission datapath is designed to provide a high throughput rate transmission, ensuring the frequency results are uploaded without interruption. Several experiments are performed to evaluate the designed real-time frequency measurement system, the input signal has a bandwidth of 4 GHz, and the repetition rate of frames is 22 MHz. Experimental results show that the frequency of the signal can be measured at a high sampling rate of 20 GSPS, and the frequency precision is better than 1 MHz.

eess.SP↗
arXiv2023

RFD-ECNet: Extreme Underwater Image Compression with Reference to Feature Dictionar

Thriving underwater applications demand efficient extreme compression technology to realize the transmission of underwater images (UWIs) in very narrow underwater bandwidth. However, existing image compression methods achieve inferior performance on UWIs because they do not consider the characteristics of UWIs: (1) Multifarious underwater styles of color shift and distance-dependent clarity, caused by the unique underwater physical imaging; (2) Massive redundancy between different UWIs, caused by the fact that different UWIs contain several common ocean objects, which have plenty of similarities in structures and semantics. To remove redundancy among UWIs, we first construct an exhaustive underwater multi-scale feature dictionary to provide coarse-to-fine reference features for UWI compression. Subsequently, an extreme UWI compression network with reference to the feature dictionary (RFD-ECNet) is creatively proposed, which utilizes feature match and reference feature variant to significantly remove redundancy among UWIs. To align the multifarious underwater styles and improve the accuracy of feature match, an underwater style normalized block (USNB) is proposed, which utilizes underwater physical priors extracted from the underwater physical imaging model to normalize the underwater styles of dictionary features toward the input. Moreover, a reference feature variant module (RFVM) is designed to adaptively morph the reference features, improving the similarity between the reference and input features. Experimental results on four UWI datasets show that our RFD-ECNet is the first work that achieves a significant BD-rate saving of 31% over the most advanced VVC.

cs.CV↗
arXiv2023

Many-body physics of spontaneously broken higher-rank symmetry: from fractonic superfluids to dipolar Hubbard model

Fractonic superfluids are exotic phases of matter in which bosons are subject to mobility constraints, resulting in features beyond those of conventional superfluids. These exotic phases arise from the spontaneous breaking of higher-rank symmetry (HRS) in many-body systems with higher-moment conservation, such as dipoles, quadrupoles, and angular moments. The aim of this paper is to introduce exciting developments on the theory of spontaneous symmetry breaking in such systems, which we refer to as ``many-fracton systems''. More specifically, we introduce exciting progress on general aspects of HRS, minimal model construction, realization of symmetry-breaking ground states, order parameter, off-diagonal long-range order (ODLRO), Noether currents with continuity equations, Gross-Pitaevskii equations, quantum fluctuations, Goldstone modes, specific heat, generalized Mermin-Wagner theorem, critical current, Landau criterion, symmetry defects, and Kosterlitz-Thouless (KT)-like physics, hydrodynamics, and dipolar Hubbard model realization. This paper is concluded with several future directions.

cond-mat.str-el↗
arXiv2023

Continuum field theory of 3D topological orders with emergent fermions and braiding statistics

Universal topological data of topologically ordered phases can be captured by topological quantum field theory in continuous space time by taking the limit of low energies and long wavelengths. While previous continuum field-theoretical studies of topological orders in $3$D real space focus on either self-statistics, braiding statistics, shrinking rules, fusion rules or quantum dimensions, it is yet to systematically put all topological data together in a unified continuum field-theoretical framework. Here, we construct the topological $BF$ field theory with twisted terms (e.g., $AAdA$ and $AAB$) as well as a $K$-matrix $BB$ term, in order to simultaneously explore all such topological data and reach anomaly-free topological orders. Following the spirit of the famous $K$-matrix Chern-Simons theory of $2$D topological orders, we present general formulas and systematically show how the $K$-matrix $BB$ term confines topological excitations, and how self-statistics of particles is transmuted between bosonic one and fermionic one. In order to reach anomaly-free topological orders, we explore, within the present continuum field-theoretical framework, how the principle of gauge invariance fundamentally influences possible realizations of topological data. More concretely, we present the topological actions of (i) particle-loop braidings with emergent fermions, (ii) multiloop braidings with emergent fermions, and (iii) Borromean-Rings braidings with emergent fermions, and calculate their universal topological data. Together with the previous efforts, our work paves the way toward a more systematic and complete continuum field-theoretical analysis of exotic topological properties of $3$D topological orders. Several interesting future directions are also discussed.

cond-mat.str-el↗
arXiv2023

Topological fracton quantum phase transitions by tuning exact tensor network states

Gapped fracton phases of matter generalize the concept of topological order and broaden our fundamental understanding of entanglement in quantum many-body systems. However, their analytical or numerical description beyond exactly solvable models remains a formidable challenge. Here we employ an exact 3D quantum tensor-network approach that allows us to study a $\mathbb{Z}_N$ generalization of the prototypical X cube fracton model and its quantum phase transitions between distinct topological states via fully tractable wavefunction deformations. We map the (deformed) quantum states exactly to a combination of a classical lattice gauge theory and a plaquette clock model, and employ numerical techniques to calculate various entanglement order parameters. For the $\mathbb{Z}_N$ model we find a family of (weakly) first-order fracton confinement transitions that in the limit of $N\to\infty$ converge to a continuous phase transition beyond the Landau-Ginzburg-Wilson paradigm. We also discover a line of 3D conformal quantum critical points (with critical magnetic flux loop fluctuations) which, in the $N\to\infty$ limit, appears to coexist with a gapless deconfined fracton state.

cond-mat.str-el↗
arXiv2023

Stimulative Training++: Go Beyond The Performance Limits of Residual Networks

Residual networks have shown great success and become indispensable in recent deep neural network models. In this work, we aim to re-investigate the training process of residual networks from a novel social psychology perspective of loafing, and further propose a new training scheme as well as three improved strategies for boosting residual networks beyond their performance limits. Previous research has suggested that residual networks can be considered as ensembles of shallow networks, which implies that the final performance of a residual network is influenced by a group of subnetworks. We identify a previously overlooked problem that is analogous to social loafing, where subnetworks within a residual network are prone to exert less effort when working as part of a group compared to working alone. We define this problem as \textit{network loafing}. Similar to the decreased individual productivity and overall performance as demonstrated in society, network loafing inevitably causes sub-par performance. Inspired by solutions from social psychology, we first propose a novel training scheme called stimulative training, which randomly samples a residual subnetwork and calculates the KL divergence loss between the sampled subnetwork and the given residual network for extra supervision. In order to unleash the potential of stimulative training, we further propose three simple-yet-effective strategies, including a novel KL- loss that only aligns the network logits direction, random smaller inputs for subnetworks, and inter-stage sampling rules. Comprehensive experiments and analysis verify the effectiveness of stimulative training as well as its three improved strategies.

cs.LG↗
arXiv2023

Detecting bulk and edge exceptional points in non-Hermitian systems through generalized Petermann factors

Non-orthogonality in non-Hermitian quantum systems gives rise to tremendous exotic quantum phenomena, which can be fundamentally traced back to non-unitarity and is much more fundamental and universal than complex energy spectrum. In this paper, we introduce an interesting quantity (denoted as $η$) as a new variant of the Petermann factor to directly and efficiently measure non-unitarity and the associated non-Hermitian physics. By tuning the model parameters of underlying non-Hermitian systems, we find that the discontinuity of both $η$ and its first-order derivative (denoted as $\partial η$) pronouncedly captures rich physics that is fundamentally caused by non-unitarity. More concretely, in the 1D non-Hermitian topological systems, two mutually orthogonal edge states that are respectively localized on two boundaries become non-orthogonal in the vicinity of discontinuity of $η$ as a function of the model parameter, which is dubbed ``edge state transition''. Through theoretical analysis, we identify that the appearance of edge state transition indicates the existence of exceptional points~(EPs) in topological edge states. Regarding the discontinuity of $\partialη$, we investigate a two-level non-Hermitian model and establish a connection between the points of discontinuity of $\partial η$ and EPs of bulk states. By studying this connection in more general lattice models, we find that some models have discontinuity of $\partialη$, implying the existence of EPs in bulk states.

quant-ph↗
arXiv2023

Induced-photorefractive attack against Quantum Key Distribution

Lithium niobate (LiNbO3, LN) devices play critical roles in quantum information processing. However, for special applications like quantum key distribution (QKD), the characteristics of materials and devices and their impact on practical systems must be intensively inquired. For the first time, we reveal that the photorefractive effect in LN can be utilized as a potential loophole to carry out malicious attacks by the eavesdroppers. We take a commercial LN-based variable optical attenuator as an example to demonstrate the method we named Induced-photorefractive attack (IPA) and propose two techniques to enable controllable attacks. Our results show that eavesdroppers can fulfill an efficient source-side attack by injecting an optimized irradiation beam with only several nanowatts, which is realistic when accessing commercial fiber channels. These measure and techniques can be employed for all individual and on-chip LN devices and initially explored a new security branch for system design and standardization of real-life QKD.

quant-ph↗
arXiv2023

A2S-NAS: Asymmetric Spectral-Spatial Neural Architecture Search For Hyperspectral Image Classification

Existing deep learning-based hyperspectral image (HSI) classification works still suffer from the limitation of the fixed-sized receptive field, leading to difficulties in distinctive spectral-spatial features for ground objects with various sizes and arbitrary shapes. Meanwhile, plenty of previous works ignore asymmetric spectral-spatial dimensions in HSI. To address the above issues, we propose a multi-stage search architecture in order to overcome asymmetric spectral-spatial dimensions and capture significant features. First, the asymmetric pooling on the spectral-spatial dimension maximally retains the essential features of HSI. Then, the 3D convolution with a selectable range of receptive fields overcomes the constraints of fixed-sized convolution kernels. Finally, we extend these two searchable operations to different layers of each stage to build the final architecture. Extensive experiments are conducted on two challenging HSI benchmarks including Indian Pines and Houston University, and results demonstrate the effectiveness of the proposed method with superior performance compared with the related works.

cs.CV↗
arXiv2023

Spectrally tunable ultrashort monochromatized extreme ultraviolet pulses at 100 kHz

We present the experimental realization of spectrally tunable, ultrashort, quasi-monochromatic extreme ultraviolet (XUV) pulses generated at 100 kHz repetition rate in a user-oriented gas high harmonic generation (GHHG) beamline of the Extreme Light Infrastructure - Attosecond Light Pulse Source (ELI ALPS) facility. Versatile spectral and temporal shaping of the XUV pulses are accomplished with a double-grating, time-delay compensated monochromator accommodating the two composing stages in a novel, asymmetrical geometry. This configuration supports the achievement of high monochromatic XUV flux (2.8+/-0.9*1e10 photons/s at 39.7 eV selected with 700 meV FWHM bandwidth) combined with ultrashort pulse duration (4.0+/-0.2 fs using 12.1+/-0.6 fs driving pulses) and small spot size (sub-100 um). Focusability, spectral bandwidth, and overall photon flux of the produced radiation were investigated covering a wide range of instrumental configurations. Moreover, complete temporal (intensity and phase) characterization of the few-femtosecond monochromatic XUV pulses - a goal that is difficult to achieve by conventional reconstruction techniques - has been realized using ptychographic algorithm on experimentally recorded XUV-IR pump-probe traces. The presented results contribute to in-situ, time-resolved experiments accessing direct information on the electronic structure dynamics of novel target materials.

physics.optics↗
arXiv2023

JNDMix: JND-Based Data Augmentation for No-reference Image Quality Assessment

Despite substantial progress in no-reference image quality assessment (NR-IQA), previous training models often suffer from over-fitting due to the limited scale of used datasets, resulting in model performance bottlenecks. To tackle this challenge, we explore the potential of leveraging data augmentation to improve data efficiency and enhance model robustness. However, most existing data augmentation methods incur a serious issue, namely that it alters the image quality and leads to training images mismatching with their original labels. Additionally, although only a few data augmentation methods are available for NR-IQA task, their ability to enrich dataset diversity is still insufficient. To address these issues, we propose a effective and general data augmentation based on just noticeable difference (JND) noise mixing for NR-IQA task, named JNDMix. In detail, we randomly inject the JND noise, imperceptible to the human visual system (HVS), into the training image without any adjustment to its label. Extensive experiments demonstrate that JNDMix significantly improves the performance and data efficiency of various state-of-the-art NR-IQA models and the commonly used baseline models, as well as the generalization ability. More importantly, JNDMix facilitates MANIQA to achieve the state-of-the-art performance on LIVEC and KonIQ-10k.

cs.CV↗
arXiv2023

$β$-DARTS++: Bi-level Regularization for Proxy-robust Differentiable Architecture Search

Neural Architecture Search has attracted increasing attention in recent years. Among them, differential NAS approaches such as DARTS, have gained popularity for the search efficiency. However, they still suffer from three main issues, that are, the weak stability due to the performance collapse, the poor generalization ability of the searched architectures, and the inferior robustness to different kinds of proxies. To solve the stability and generalization problems, a simple-but-effective regularization method, termed as Beta-Decay, is proposed to regularize the DARTS-based NAS searching process (i.e., $β$-DARTS). Specifically, Beta-Decay regularization can impose constraints to keep the value and variance of activated architecture parameters from being too large, thereby ensuring fair competition among architecture parameters and making the supernet less sensitive to the impact of input on the operation set. In-depth theoretical analyses on how it works and why it works are provided. Comprehensive experiments validate that Beta-Decay regularization can help to stabilize the searching process and makes the searched network more transferable across different datasets. To address the robustness problem, we first benchmark different NAS methods under a wide range of proxy data, proxy channels, proxy layers and proxy epochs, since the robustness of NAS under different kinds of proxies has not been explored before. We then conclude some interesting findings and find that $β$-DARTS always achieves the best result among all compared NAS methods under almost all proxies. We further introduce the novel flooding regularization to the weight optimization of $β$-DARTS (i.e., Bi-level regularization), and experimentally and theoretically verify its effectiveness for improving the proxy robustness of differentiable NAS.

cs.LG↗
arXiv2022

Hierarchy of Entanglement Renormalization and Long-Range Entangled States

As a quantum-informative window into quantum many-body physics, the concept and application of entanglement renormalization group (ERG) have been playing a vital role in the study of novel quantum phases of matter, especially long-range entangled (LRE) states in topologically ordered systems. For instance, by recursively applying local unitaries as well as adding/removing qubits that form product states, the 2D toric code ground states, i.e., fixed point of Z_2 topological order, are efficiently coarse-grained with respect to the system size. As a further improvement, the addition/removal of 2D toric codes into/from the ground states of the 3D X-cube model, is shown to be indispensable and remarkably leads to well-defined fixed points of a large class of fracton orders that are non-liquid-like. Here, we present a substantially unified ERG framework in which general degrees of freedom are allowed to be recursively added/removed. Specifically, we establish an exotic hierarchy of ERG and LRE states in Pauli stabilizer codes, where the 2D toric code and 3D X-cube models are naturally included. In the hierarchy, LRE states like 3D X-cube and 3D toric code ground states can be added/removed in ERG processes of more complex LRE states. In this way, a large group of Pauli stabilizer codes are categorized into a series of ``state towers''; with each tower, in addition to local unitaries including CNOT gates, lower LRE states of level-$n$ are added/removed in the level-$n$ ERG process of an upper LRE state of level-$(n+1)$, connecting LRE states of different levels and unveiling complex relations among LRE states. As future directions, we expect this hierarchy can be applied to more general LRE states, leading to a unified ERG scenario of LRE states and exact tensor-network representations in the form of more generalized branching MERA.

quant-ph↗
arXiv2022

Detecting Subsystem Symmetry Protected Topological Order Through Strange Correlators

We employ strange correlators to detect 2D subsystem symmetry-protected topological (SSPT) phases which are nontrivial topological phases protected by subsystem symmetries. Specifically, we analytically construct efficient strange correlators in the 2D cluster model in the presence of a uniform magnetic field and then perform the projector Quantum Monte Carlo simulation within the quantum annealing scheme. We find that strange correlators show the long-range correlation in the SSPT phase, from which we define strange order parameters to characterize the topological phase transition between the SSPT phase at low fields and the trivial paramagnetic phase at high fields. Thus, the detection of the fully localized zero modes on the 1D physical boundary of SSPT phase has been transformed into the bulk correlation measurement about the local operators with the periodic boundary condition. We also find interesting spatial anisotropy of a strange correlator, which can be intrinsically traced back to the nature of spatial anisotropy of subsystem symmetries that protect SSPT order in the 2D cluster model. By simulating strange correlators, we, therefore, provide the first unbiased large-scale quantum Monte Carlo simulation on the easy and efficient detection in the SSPT phase and open the avenue of the investigation of the subtle yet fundamental nature of the novel interacting topological phases.

cond-mat.str-el↗
arXiv2022

Hierarchical Proliferation of Higher-Rank Symmetry Defects in Fractonic Superfluids

Symmetry defects, e.g., vortices in conventional superfluids, play a critical role in a complete description of symmetry-breaking phases. In this paper, we develop the theory of symmetry defects in fractonic superfluids, i.e., spontaneously higher-rank symmetry (HRS) breaking phases. By Noether's theorem, HRS is associated with the conservation law of higher moments, e.g. dipoles, quadrupoles, and angular moments. We establish finite-temperature phase diagrams by identifying a series of topological phase transitions via the renormalization group flow equations and Debye-Hückel approximation. Accordingly, a series of Kosterlitz-Thouless topological transitions are found to occur successively at different temperatures, which are triggered by proliferation of defects, defect bound states, and so on. Such a \emph{hierarchical proliferation} brings rich phase structures. Meanwhile, a screening effect from sufficiently high density of defect bound states leads to instability and collapse of the intermediate temperature phases, which further enriches the phase diagrams. For concreteness, we consider a fractonic superfluid in which ``angular moments'' are conserved. We then present the general theory, in which other types of HRS can be analyzed in a similar manner. Further directions are present at the end of the paper.

cond-mat.str-el↗
arXiv2022

Non-Abelian Fusion, Shrinking and Quantum Dimensions of Abelian Gauge Fluxes

Braiding and fusion rules of topological excitations are indispensable topological invariants in topological quantum computation and topological orders. While excitations in 2D are always particle-like anyons, those in 3D incorporate not only particles but also loops -- spatially nonlocal objects -- making it novel and challenging to study topological invariants in higher dimensions. While 2D fusion rules have been well understood from bulk Chern-Simons field theory and edge conformal field theory, it is yet to be thoroughly explored for 3D fusion rules from higher dimensional bulk topological field theory. Here, we perform a field-theoretical study on (i) how loops that carry Abelian gauge fluxes fuse and (ii) how loops are shrunk into particles in the path integral, which generates fusion rules, loop-shrinking rules, and descendent invariants, e.g., quantum dimensions. We first assign a gauge-invariant Wilson operator to each excitation and determine the number of distinct excitations through equivalence classes of Wilson operators. Then, we adiabatically shift two Wilson operators together to observe how they fuse and are split in the path integral; despite the Abelian nature of the gauge fluxes carried by loops, their fusions may be of non-Abelian nature. Meanwhile, we adiabatically deform world-sheets of unknotted loops into world-lines and examine the shrinking outcomes; we find that the resulting loop-shrinking rules are algebraically consistent to fusion rules. Interestingly, fusing a pair of loop and anti-loop may generate multiple vacua, but fusing a pair of anyon and anti-anyon in 2D has one vacuum only. By establishing a field-theoretical ground for fusion and shrinking in 3D, this work leaves intriguing directions, e.g., symmetry enrichment, quantum gates, and physics of braided monoidal 2-category of 2-group.

cond-mat.str-el↗
arXiv2022

Feature Reconstruction Attacks and Countermeasures of DNN training in Vertical Federated Learning

Federated learning (FL) has increasingly been deployed, in its vertical form, among organizations to facilitate secure collaborative training over siloed data. In vertical FL (VFL), participants hold disjoint features of the same set of sample instances. Among them, only one has labels. This participant, known as the active party, initiates the training and interacts with the other participants, known as the passive parties. Despite the increasing adoption of VFL, it remains largely unknown if and how the active party can extract feature data from the passive party, especially when training deep neural network (DNN) models. This paper makes the first attempt to study the feature security problem of DNN training in VFL. We consider a DNN model partitioned between active and passive parties, where the latter only holds a subset of the input layer and exhibits some categorical features of binary values. Using a reduction from the Exact Cover problem, we prove that reconstructing those binary features is NP-hard. Through analysis, we demonstrate that, unless the feature dimension is exceedingly large, it remains feasible, both theoretically and practically, to launch a reconstruction attack with an efficient search-based algorithm that prevails over current feature protection techniques. To address this problem, we develop a novel feature protection scheme against the reconstruction attack that effectively misleads the search to some pre-specified random values. With an extensive set of experiments, we show that our protection scheme sustains the feature reconstruction attack in various VFL applications at no expense of accuracy loss.

cs.LG↗
arXiv2022

Stimulative Training of Residual Networks: A Social Psychology Perspective of Loafing

Residual networks have shown great success and become indispensable in today's deep models. In this work, we aim to re-investigate the training process of residual networks from a novel social psychology perspective of loafing, and further propose a new training strategy to strengthen the performance of residual networks. As residual networks can be viewed as ensembles of relatively shallow networks (i.e., \textit{unraveled view}) in prior works, we also start from such view and consider that the final performance of a residual network is co-determined by a group of sub-networks. Inspired by the social loafing problem of social psychology, we find that residual networks invariably suffer from similar problem, where sub-networks in a residual network are prone to exert less effort when working as part of the group compared to working alone. We define this previously overlooked problem as \textit{network loafing}. As social loafing will ultimately cause the low individual productivity and the reduced overall performance, network loafing will also hinder the performance of a given residual network and its sub-networks. Referring to the solutions of social psychology, we propose \textit{stimulative training}, which randomly samples a residual sub-network and calculates the KL-divergence loss between the sampled sub-network and the given residual network, to act as extra supervision for sub-networks and make the overall goal consistent. Comprehensive empirical results and theoretical analyses verify that stimulative training can well handle the loafing problem, and improve the performance of a residual network by improving the performance of its sub-networks. The code is available at https://github.com/Sunshine-Ye/NIPS22-ST .

cs.CV↗