arXiv subjects

Lin Zhang

Publications and source records attributed to Lin Zhang.

At least 289 records · Page 16Linked to original sources

arXiv2021

Bayesian functional graphical models

We develop a Bayesian graphical modeling framework for functional data for correlated multivariate random variables observed over a continuous domain. Our method leads to graphical Markov models for functional data which allows the graphs to vary over the functional domain. The model involves estimation of graphical models that evolve functionally in a nonparametric fashion while accounting for within-functional correlations and borrowing strength across functional positions so contiguous locations are encouraged but not forced to have similar graph structure and edge strength. We utilize a strategy that combines nonparametric basis function modeling with modified Bayesian graphical regularization techniques, which induces a new class of hypoexponential normal scale mixture distributions that not only leads to adaptively shrunken estimators of the conditional cross-covariance but also facilitates a thorough theoretical investigation of the shrinkage properties. Our approach scales up to large functional datasets collected on a fine grid. We show through simulations and real data analysis that the Bayesian functional graphical model can efficiently reconstruct the functionally-evolving graphical models by accounting for within-function correlations.

stat.ME↗

arXiv2021

A Credibility-aware Swarm-Federated Deep Learning Framework in Internet of Vehicles

Federated Deep Learning (FDL) is helping to realize distributed machine learning in the Internet of Vehicles (IoV). However, FDL's global model needs multiple clients to upload learning model parameters, thus still existing unavoidable communication overhead and data privacy risks. The recently proposed Swarm Learning (SL) provides a decentralized machine-learning approach uniting edge computing and blockchain-based coordination without the need for a central coordinator. This paper proposes a Swarm-Federated Deep Learning framework in the IoV system (IoV-SFDL) that integrates SL into the FDL framework. The IoV-SFDL organizes vehicles to generate local SL models with adjacent vehicles based on the blockchain empowered SL, then aggregates the global FDL model among different SL groups with a proposed credibility weights prediction algorithm. Extensive experimental results demonstrate that compared with the baseline frameworks, the proposed IoV-SFDL framework achieves a 16.72% reduction in edge-to-global communication overhead while improving about 5.02% in model performance with the same training iterations.

eess.SY↗

arXiv2021

Edge of chaos as a guiding principle for modern neural network training

The success of deep neural networks in real-world problems has prompted many attempts to explain their training dynamics and generalization performance, but more guiding principles for the training of neural networks are still needed. Motivated by the edge of chaos principle behind the optimal performance of neural networks, we study the role of various hyperparameters in modern neural network training algorithms in terms of the order-chaos phase diagram. In particular, we study a fully analytical feedforward neural network trained on the widely adopted Fashion-MNIST dataset, and study the dynamics associated with the hyperparameters in back-propagation during the training process. We find that for the basic algorithm of stochastic gradient descent with momentum, in the range around the commonly used hyperparameter values, clear scaling relations are present with respect to the training time during the ordered phase in the phase diagram, and the model's optimal generalization power at the edge of chaos is similar across different training parameter combinations. In the chaotic phase, the same scaling no longer exists. The scaling allows us to choose the training parameters to achieve faster training without sacrificing performance. In addition, we find that the commonly used model regularization method - weight decay - effectively pushes the model towards the ordered phase to achieve better performance. Leveraging on this fact and the scaling relations in the other hyperparameters, we derived a principled guideline for hyperparameter determination, such that the model can achieve optimal performance by saturating it at the edge of chaos. Demonstrated on this simple neural network model and training algorithm, our work improves the understanding of neural network training dynamics, and can potentially be extended to guiding principles of more complex model architectures and algorithms.

cs.LG↗

arXiv2021

Accelerating Distributed K-FAC with Smart Parallelism of Computing and Communication Tasks

Distributed training with synchronous stochastic gradient descent (SGD) on GPU clusters has been widely used to accelerate the training process of deep models. However, SGD only utilizes the first-order gradient in model parameter updates, which may take days or weeks. Recent studies have successfully exploited approximate second-order information to speed up the training process, in which the Kronecker-Factored Approximate Curvature (KFAC) emerges as one of the most efficient approximation algorithms for training deep models. Yet, when leveraging GPU clusters to train models with distributed KFAC (D-KFAC), it incurs extensive computation as well as introduces extra communications during each iteration. In this work, we propose D-KFAC (SPD-KFAC) with smart parallelism of computing and communication tasks to reduce the iteration time. Specifically, 1) we first characterize the performance bottlenecks of D-KFAC, 2) we design and implement a pipelining mechanism for Kronecker factors computation and communication with dynamic tensor fusion, and 3) we develop a load balancing placement for inverting multiple matrices on GPU clusters. We conduct real-world experiments on a 64-GPU cluster with 100Gb/s InfiniBand interconnect. Experimental results show that our proposed SPD-KFAC training scheme can achieve 10%-35% improvement over state-of-the-art algorithms.

cs.DC↗

arXiv2021

Spectra of glueballs and oddballs and the equation of state from holographic QCD

We study the spectra of two-gluon glueballs and three-gluon oddballs and corresponding equation of state in $5$-dimensional deformed holographic QCD models in the graviton-dilaton system, where the metric, the dilaton field and dilaton potential are self-consistently solved from each other through the Einstein field equations and the equation of motion of the dilaton field. We compare the models by inputting the dilaton field, inputting the deformed metric and inputting the dilaton potential, and find that with only 2 parameters, the $5$-dimensional holographic QCD model predictions on glueballs/oddballs spectra in general are in good agreement with lattice results except two oddballs $0^{+-}$ and $2^{+-}$. From the results of glueballs/oddballs spectra at zero temperature and the equation of state at finite temperature, we observe that the model with quadratic dilaton field can simultaneously describe glueballs/oddballs spectra as well as equation of state of pure gluon system. The model with quadratic $A_{E}(z)$ can describe glueballs/oddballs spectra, but its corresponding equation of state behaves more like $N_{f}=2+1$ quark matter. These are consistent with dimension analysis at UV boundary.

hep-ph↗

arXiv2021

Temporal Graph Signal Decomposition

Temporal graph signals are multivariate time series with individual components associated with nodes of a fixed graph structure. Data of this kind arises in many domains including activity of social network users, sensor network readings over time, and time course gene expression within the interaction network of a model organism. Traditional matrix decomposition methods applied to such data fall short of exploiting structural regularities encoded in the underlying graph and also in the temporal patterns of the signal. How can we take into account such structure to obtain a succinct and interpretable representation of temporal graph signals? We propose a general, dictionary-based framework for temporal graph signal decomposition (TGSD). The key idea is to learn a low-rank, joint encoding of the data via a combination of graph and time dictionaries. We propose a highly scalable decomposition algorithm for both complete and incomplete data, and demonstrate its advantage for matrix decomposition, imputation of missing values, temporal interpolation, clustering, period estimation, and rank estimation in synthetic and real-world data ranging from traffic patterns to social media activity. Our framework achieves 28% reduction in RMSE compared to baselines for temporal interpolation when as many as 75% of the observations are missing. It scales best among baselines taking under 20 seconds on 3.5 million data points and produces the most parsimonious models. To the best of our knowledge, TGSD is the first framework to jointly model graph signals by temporal and graph dictionaries.

cs.LG↗

arXiv2021

Gluodynamics and deconfinement phase transition under rotation from holography

We investigate rotating effect on deconfinement phase transition in an Einstein-Maxwell-Dilaton(EMD) model in bottom-up holographic QCD approach. By constructing a rotating black hole, which is supposed to be dual to rotating strongly coupled nuclear matter, we investigate the thermodynamic quantities, including entropy density, pressure, energy density, trace anomaly, sound speed and specific heat for both pure gluon system and two-flavor system under rotation. It is shown that those thermodynamic quantities would be enhanced by large angular velocity. Also, we extract the information of phase transition from those thermodynamic quantities, as well as the order parameter of deconfinement phase transition, i.e. the loop operators. It is shown that, in the $T - ω$ plane, for two-flavor case with small chemical potential, the phase transition is always crossover. The transition temperature decreases slowly with angular velocity and chemical potential. For pure gluon system with zero chemical potential, the phase transition is always first order, while at finite chemical potential a critical end point(CEP) will present in the $T - ω$ plane.

hep-ph↗

arXiv2021

An Initial Investigation for Detecting Partially Spoofed Audio

All existing databases of spoofed speech contain attack data that is spoofed in its entirety. In practice, it is entirely plausible that successful attacks can be mounted with utterances that are only partially spoofed. By definition, partially-spoofed utterances contain a mix of both spoofed and bona fide segments, which will likely degrade the performance of countermeasures trained with entirely spoofed utterances. This hypothesis raises the obvious question: 'Can we detect partially-spoofed audio?' This paper introduces a new database of partially-spoofed data, named PartialSpoof, to help address this question. This new database enables us to investigate and compare the performance of countermeasures on both utterance- and segmental- level labels. Experimental results using the utterance-level labels reveal that the reliability of countermeasures trained to detect fully-spoofed data is found to degrade substantially when tested with partially-spoofed data, whereas training on partially-spoofed data performs reliably in the case of both fully- and partially-spoofed utterances. Additional experiments using segmental-level labels show that spotting injected spoofed segments included in an utterance is a much more challenging task even if the latest countermeasure models are used.

eess.AS↗

arXiv2021

A One-Shot Texture-Perceiving Generative Adversarial Network for Unsupervised Surface Inspection

Visual surface inspection is a challenging task owing to the highly diverse appearance of target surfaces and defective regions. Previous attempts heavily rely on vast quantities of training examples with manual annotation. However, in some practical cases, it is difficult to obtain a large number of samples for inspection. To combat it, we propose a hierarchical texture-perceiving generative adversarial network (HTP-GAN) that is learned from the one-shot normal image in an unsupervised scheme. Specifically, the HTP-GAN contains a pyramid of convolutional GANs that can capture the global structure and fine-grained representation of an image simultaneously. This innovation helps distinguishing defective surface regions from normal ones. In addition, in the discriminator, a texture-perceiving module is devised to capture the spatially invariant representation of normal image via directional convolutions, making it more sensitive to defective areas. Experiments on a variety of datasets consistently demonstrate the effectiveness of our method.

cs.CV↗

arXiv2021

Andreev reflection adjusting in the multi-terminal device with the kink states

At the domain wall between two regions with the opposite Chern number, there should be the one-dimensional chiral states, which are called as the kink states. The kink states are robust for the lattice deformations. We design a multi-terminal device with the kink states to study the local Andreev reflection and the crossed Andreev reflection. The coefficient of the crossed Andreev reflection can reach 1 in the four-terminal device. Besides adjusting the phase difference between superconductors, the local Andreev reflection and the crossed Andreev reflection can be controlled by changing the on-site energy and the stagger energy in the four-terminal device. Our results give some new ideas to design the quantum device in the future.

cond-mat.mes-hall↗

arXiv2021

Nonreciprocal optical solitons in a spinning Kerr resonator

We propose a spinning nonlinear resonator as an experimentally accessible platform to achieve nonreciprocal control of optical solitons. Nonreciprocity here results from the relativistic Sagnac-Fizeau optical drag effect, which is different for pump fields propagating in the spinning direction or in the direction opposite to it. We show that in a spinning Kerr resonator, different soliton states appear for the input fields in different directions. These nonreciprocal solitons are more stable against losses induced by inter-modal coupling between clockwise and counterclockwise modes of the resonator. Our work builds a bridge between nonreciprocal physics and soliton science, providing a promising route towards achieving soliton-wave optical isolators and one-way soliton communications.

physics.optics↗

arXiv2021

Quantum dynamical characterization and simulation of topological phases with high-order band inversion surfaces

How to characterize topological quantum phases is a fundamental issue in the broad field of topological matter. From a dimension reduction approach, we propose the concept of high-order band inversion surfaces (BISs) which enable the optimal schemes to characterize equilibrium topological phases by far-from-equilibrium quantum dynamics, and further report the experimental simulation. We show that characterization of a d-dimensional (dD) topological phase can be reduced to lower-dimensional topological invariants in the high-order BISs, of which the nth-order BIS is a (d-n)D interface in momentum space. In quenching the system from trivial phase to topological regime, we unveil a high-order dynamical bulk-surface correspondence that the quantum dynamics exhibits nontrivial topological pattern in arbitrary nth-order BISs, which universally corresponds to and so characterizes the equilibrium topological phase of the post-quench Hamiltonian. This high-order dynamical bulk-surface correspondence provides new and optimal dynamical schemes with fundamental advantages to simulate and detect topological states, in which through the highest-order BISs that are of zero dimension, the detection of topological phase relies on only minimal measurements. We experimentally build up a quantum simulator with spin qubits to investigate a 3D chiral topological insulator through emulating each momentum one by one and measure the high-order dynamical bulk-surface correspondence, with the advantages of topological characterization via highest-order BISs being demonstrated.

cond-mat.mes-hall↗

arXiv2021

Density-Aware Federated Imitation Learning for Connected and Automated Vehicles with Unsignalized Intersection

Intelligent Transportation System (ITS) has become one of the essential components in Industry 4.0. As one of the critical indicators of ITS, efficiency has attracted wide attention from researchers. However, the next generation of urban traffic carried by multiple transport service providers may prohibit the raw data interaction among multiple regions for privacy reasons, easily ignored in the existing research. This paper puts forward a federated learning-based vehicle control framework to solve the above problem, including interactors, trainers, and an aggregator. In addition, the density-aware model aggregation method is utilized in this framework to improve vehicle control. What is more, to promote the performance of the end-to-end learning algorithm in the safety aspect, this paper proposes an imitation learning algorithm, which can obtain collision avoidance capabilities from a set of collision avoidance rules. Furthermore, a loss-aware experience selection strategy is also explored, reducing the communication overhead between the interactors and the trainers via extra computing. Finally, the experiment results demonstrate that the proposed imitation learning algorithm obtains the ability to avoid collisions and reduces discomfort by 55.71%. Besides, density-aware model aggregation can further reduce discomfort by 41.37%, and the experience selection scheme can reduce the communication overhead by 12.80% while ensuring model convergence.

cs.MA↗

arXiv2021

Causal Discovery of Flight Service Process Based on Event Sequence

The development of the civil aviation industry has continuously increased the requirements for the efficiency of airport ground support services. In the existing ground support research, there has not yet been a process model that directly obtains support from the ground support log to study the causal relationship between service nodes and flight delays. Most ground support studies mainly use machine learning methods to predict flight delays, and the flight support model they are based on is an ideal model. The study did not conduct an in-depth study of the causal mechanism behind the ground support link and did not reveal the true cause of flight delays. Therefore, there is a certain deviation in the prediction of flight delays by machine learning, and there is a certain deviation between the ideal model based on the research and the actual service process. Therefore, it is of practical significance to obtain the process model from the guarantee log and analyze its causality. However, the existing process causal factor discovery methods only do certain research when the assumption of causal sufficiency is established and does not consider the existence of latent variables. Therefore, this article proposes a framework to realize the discovery of process causal factors without assuming causal sufficiency. The optimized fuzzy mining process model is used as the service benchmark model, and the local causal discovery algorithm is used to discover the causal factors. Under this framework, this paper proposes a new Markov blanket discovery algorithm that does not assume causal sufficiency to discover causal factors and uses benchmark data sets for testing. Finally, the actual flight service data is used.

cs.LG↗

arXiv2021

Universal topological quench dynamics: Altland-Zirnbauer tenfold classes

Topological phases of the famous Altland-Zirnbauer (AZ) tenfold classes are defined on the equilibrium ground states. Whether such equilibrium topological phases have universal correspondence to far-from-equilibrium quantum dynamics is a fundamental issue of both theoretical and experimental importance. Here we uncover the universal topological quench dynamics linking to the equilibrium topological phases for the complete AZ tenfold classes, with a general framework being established. We show a fundamental result that a $d$-dimensional topological phase of the tenfold class, with an integer invariant or $\mathbb{Z}_{2}$ index defined on high symmetry momenta, is generically characterized by topology reduced to the highest-order band-inversion surfaces located at arbitrary discrete momenta of Brillouin zone. Such dimension-reduced topology is further captured by universal topological patterns emerging in far-from-equilibrium quantum dynamics by quenching the system from trivial phase to the topological regime, rendering the dynamical hallmark of the equilibrium topological phase. This work establishes a universal dynamical characterization for the complete AZ symmetry classes of topological phases, which has broad applications in theory and experiment.

cond-mat.mes-hall↗

arXiv2021

A Dynamics Perspective of Pursuit-Evasion Games of Intelligent Agents with the Ability to Learn

Pursuit-evasion games are ubiquitous in nature and in an artificial world. In nature, pursuer(s) and evader(s) are intelligent agents that can learn from experience, and dynamics (i.e., Newtonian or Lagrangian) is vital for the pursuer and the evader in some scenarios. To this end, this paper addresses the pursuit-evasion game of intelligent agents from the perspective of dynamics. A bio-inspired dynamics formulation of a pursuit-evasion game and baseline pursuit and evasion strategies are introduced at first. Then, reinforcement learning techniques are used to mimic the ability of intelligent agents to learn from experience. Based on the dynamics formulation and reinforcement learning techniques, the effects of improving both pursuit and evasion strategies based on experience on pursuit-evasion games are investigated at two levels 1) individual runs and 2) ranges of the parameters of pursuit-evasion games. Results of the investigation are consistent with nature observations and the natural law - survival of the fittest. More importantly, with respect to the result of a pursuit-evasion game of agents with baseline strategies, this study achieves a different result. It is shown that, in a pursuit-evasion game with a dynamics formulation, an evader is not able to escape from a slightly faster pursuer with an effective learned pursuit strategy, based on agile maneuvers and an effective learned evasion strategy.

eess.SY↗

arXiv2021

Implementing conventional and unconventional nonadiabatic geometric quantum gates via SU(2) transformations

We propose a simple but versatile protocol to engineer time-dependent Hamiltonians inversely for geometric quantum computation. By utilizing SU(2) transformation, a speedup goal on gate operation is achieved with more freedom to design the control parameters. As an application, this protocol enables the conventional and unconventional nonadiabatic geometric quantum gates with desired evolution paths by controlling the microwave pulses in the diamond nitrogen-vacancy center system. We show that the inversely designed Hamiltonian can fulfill the geometric gate with more economical evolution time and further reduces the influence of the environment noise on gate fidelity.

quant-ph↗

arXiv2021

A comparison of two approaches for measuring interdisciplinary research output: the disciplinary diversity of authors vs the disciplinary diversity of the reference list

This study investigates the convergence of two bibliometric approaches to the measurement of interdisciplinary research: one based on analyzing disciplinary diversity in the reference list of publications, the other based on the disciplinary diversity of authors of publications. In particular we measure the variety, balance, disparity and integrated diversity index of, respectively, single-author, multi-author single-field, and multi-author multi-field publications. We find that, in general, the diversity of the reference list grows with the number of fields reflected in a paper's authors' list and, to a lesser extent, with the number of authors being equal the number of fields. Further, we find that when fields belonging to different disciplines are reflected in the authors' list, the disparity in the reference list is higher than in the case of fields belonging to the same discipline. However, this general tendency varies across disciplines, and noticeable exceptions are found at individual paper level.

cs.DL↗