arXiv subjects

Lin Chen

Publications and source records attributed to Lin Chen.

At least 379 records · Page 21Linked to original sources

arXiv2021

An extensive empirical study of inconsistent labels in multi-version-project defect data sets

The label quality of defect data sets has a direct influence on the reliability of defect prediction models. In this study, for multi-version-project defect data sets, we propose an approach to automatically detecting instances with inconsistent labels (i.e. the phenomena of instances having the same source code but different labels over multiple versions of a software project) and understand their influence on the evaluation and interpretation of defect prediction models. Based on five multi-version-project defect data sets (either widely used or the most up-to-date in the literature) collected by diverse approaches, we find that: (1) most versions in the investigated defect data sets contain inconsistent labels with varying degrees; (2) the existence of inconsistent labels in a training data set may considerably change the prediction performance of a defect prediction model as well as can lead to the identification of substantially different true defective modules; and (3) the importance ranking of independent variables in a defect prediction model can be substantially shifted due to the existence of inconsistent labels. The above findings reveal that inconsistent labels in defect data sets can profoundly change the prediction ability and interpretation of a defect prediction model. Therefore, we strongly suggest that practitioners should detect and exclude inconsistent labels in defect data sets to avoid their potential negative influence on defect prediction models. What is more, it is necessary for researchers to improve existing defect label collection approaches to reduce inconsistent labels. Furthermore, there is a need to re-examine the experimental conclusions of previous studies using multi-version-project defect data sets with a high ratio of inconsistent labels.

cs.SE↗

arXiv2021

DAIL: Dataset-Aware and Invariant Learning for Face Recognition

To achieve good performance in face recognition, a large scale training dataset is usually required. A simple yet effective way to improve recognition performance is to use a dataset as large as possible by combining multiple datasets in the training. However, it is problematic and troublesome to naively combine different datasets due to two major issues. First, the same person can possibly appear in different datasets, leading to an identity overlapping issue between different datasets. Naively treating the same person as different classes in different datasets during training will affect back-propagation and generate non-representative embeddings. On the other hand, manually cleaning labels may take formidable human efforts, especially when there are millions of images and thousands of identities. Second, different datasets are collected in different situations and thus will lead to different domain distributions. Naively combining datasets will make it difficult to learn domain invariant embeddings across different datasets. In this paper, we propose DAIL: Dataset-Aware and Invariant Learning to resolve the above-mentioned issues. To solve the first issue of identity overlapping, we propose a dataset-aware loss for multi-dataset training by reducing the penalty when the same person appears in multiple datasets. This can be readily achieved with a modified softmax loss with a dataset-aware term. To solve the second issue, domain adaptation with gradient reversal layers is employed for dataset invariant learning. The proposed approach not only achieves state-of-the-art results on several commonly used face recognition validation sets, including LFW, CFP-FP, and AgeDB-30, but also shows great benefit for practical use.

cs.CV↗

arXiv2021

EDSC: An Event-Driven Smart Contract Platform

This paper presents EDSC, a novel smart contract platform design based on the event-driven execution model as opposed to the traditionally employed transaction-driven execution model. We reason that such a design is a better fit for many emerging smart contract applications and is better positioned to address the scalability and performance challenges plaguing the smart contract ecosystem. We propose EDSC's design under the Ethereum framework, and the design can be easily adapted for other existing smart contract platforms. We have conducted implementation using Ethereum client and experiments where performance modeling results show on average 2.2 to 4.6 times reduced total latency of event triggered smart contracts, which demonstrates its effectiveness for supporting contracts that demand timely execution based on events. In addition, we discuss example use cases to demonstrate the design's utility and comment on its potential security dynamics.

cs.DC↗

arXiv2020

An Optimal Mass Transport Method for Random Genetic Drift

We propose and analyze an optimal mass transport method for a random genetic drift problem driven by a Moran process under weak-selection. The continuum limit, formulated as a reaction-advection-diffusion equation known as the Kimura equation, inherits degenerate diffusion from the discrete stochastic process that conveys to the blow-up into Dirac-delta singularities hence brings great challenges to both the analytical and numerical studies. The proposed numerical method can quantitatively capture to the fullest possible extent the development of Dirac-delta singularities for genetic segregation on one hand, and preserves several sets of biologically relevant and computationally favored properties of the random genetic drift on the other. Moreover, the numerical scheme exponentially converges to the unique numerical stationary state in time at a rate independent of the mesh size up to a mesh error. Numerical evidence is given to illustrate and support these properties, and to demonstrate the spatio-temporal dynamics of random generic drift.

math.AP↗

arXiv2020

Sequential Resource Access: Theory and Algorithm

We formulate and analyze a generic sequential resource access problem arising in a variety of engineering fields, where a user disposes a number of heterogeneous computing, communication, or storage resources, each characterized by the probability of successfully executing the user's task and the related access delay and cost, and seeks an optimal access strategy to maximize her utility within a given time horizon, defined as the expected reward minus the access cost. We develop an algorithmic framework on the (near-)optimal sequential resource access strategy. We first prove that the problem of finding an optimal strategy is NP-hard in general. Given the hardness result, we present a greedy strategy implementable in linear time, and establish the closed-form sufficient condition for its optimality. We then develop a series of polynomial-time approximation algorithms achieving $(ε,δ)$-optimality, with the key component being a pruning process eliminating dominated strategies and, thus maintaining polynomial time and space overhead.

cs.NI↗

arXiv2020

On rotational flows with discontinuous vorticity beneath steady water waves near stagnation

We numerically investigate the flow structure of periodic steady water waves of fixed relative mass flux propagating on rotational flows with piece-wise constant vorticity. We show that for wave solutions along the global bifurcation diagram the stagnation point can first occur internally or at the bottom, and then again occurs at the crest with further increase in wave height. We observe that the bifurcation diagram has a new branch which is not connected to the trivial solution. Furthermore, we present an in-depth discussion of the results on pressure distributions and particle trajectories beneath large-amplitude steady waves near stagnation. We also expand on previous results concerning the amplitude and mass flux of steady water waves traveling on rotational flows with discontinuous vorticity.

physics.flu-dyn↗

arXiv2020

Block-structured Integer Programming: Can we Parameterize without the Largest Coefficient?

We consider 4-block $n$-fold integer programming, which can be written as $\max\{w\cdot x: H x=b, l\le x\le u, x\in \mathbb{Z}^{N} \}$ where the constraint matrix $H$ is composed of small submatrices $A,B,C,D$ such that the first row of $H$ is $(C,D,D,\cdots,D)$, the first column of $H$ is $(C,B,B,\cdots,B)$, the main diagonal of $H$ is $(C,A,A,\cdots,A)$, and all the other entries are $0$. The special case where $B=C=0$ is known as $n$-fold integer programming. Prior algorithmic results for 4-block $n$-fold integer programming and its special cases usually take $Δ$, the largest absolute value among entries of $H$ as part of the parameters. In this paper, we explore the possibility of getting rid of $Δ$ from parameters, i.e., we are looking for algorithms that runs polynomially in $\logΔ$. We show that, assuming $\text{P}\neq \text{NP}$, this is not possible even if $A=(1,1,Δ)$ and $B=C=0$. However, this becomes possible if $A=(1,1,\cdots,1)$ or $A\in \mathbb{Z}^{1\times 2}$, or more generally if $A\in\mathbb{Z}^{s_A\times t_A} $ where $t_A=s_A+1$ and the rank of matrix $A$ satisfies that $\text{rank}(A)=s_A$. More precisely, 1. If $A=(1,\ldots,1)\in \mathbb{Z}^{1\times t_A} $, then 4-block $n$-fold IP can be solved in $(t_A+t_B)^{O(t_A+t_B)}\cdot poly(n,\logΔ)$ time. 2. If $A\in\mathbb{Z}^{s_A\times t_A} $, $t_A=s_A+1$ and $\text{rank}(A)=s_A$, then 4-block $n$-fold IP can be solved in $(t_A+t_B)^{O(t_A+t_B)}\cdot n^{O(t_A)}\cdot poly(\logΔ)$ time; Specifically, if in addition we have $B=C=0$ (i.e., $n$-fold integer programming), then it can be solved in linear time $n\cdot poly(t_A,\log Δ)$.

math.OC↗

arXiv2020

Inertias of entanglement witnesses

Entanglement witnesses (EWs) are a fundamental tool for the detection of entanglement. We study the inertias of EWs, i.e., the triplet of the numbers of negative, zero, and positive eigenvalues respectively. We focus on the EWs constructed by the partial transposition of states with non-positive partial transposes. We provide a method to generate more inertias from a given inertia by the relevance between inertias. Based on that we exhaust all the inertias for EWs in each qubit-qudit system. We apply our results to propose a separability criterion in terms of the rank of the partial transpose of state. We also connect our results to tripartite genuinely entangled states and the classification of states with non-positive partial transposes. Additionally, the inertias of EWs constructed by X-states are clarified.

quant-ph↗

arXiv2020

Minimax Regret of Switching-Constrained Online Convex Optimization: No Phase Transition

We study the problem of switching-constrained online convex optimization (OCO), where the player has a limited number of opportunities to change her action. While the discrete analog of this online learning task has been studied extensively, previous work in the continuous setting has neither established the minimax rate nor algorithmically achieved it. In this paper, we show that $ T $-round switching-constrained OCO with fewer than $ K $ switches has a minimax regret of $ Θ(\frac{T}{\sqrt{K}}) $. In particular, it is at least $ \frac{T}{\sqrt{2K}} $ for one dimension and at least $ \frac{T}{\sqrt{K}} $ for higher dimensions. The lower bound in higher dimensions is attained by an orthogonal subspace argument. In one dimension, a novel adversarial strategy yields the lower bound of $O(\frac{T}{\sqrt{K}})$, but a precise minimax analysis including constants is more involved. To establish the tighter one-dimensional result, we introduce the \emph{fugal game} relaxation, whose minimax regret lower bounds that of switching-constrained OCO. We show that the minimax regret of the fugal game is at least $ \frac{T}{\sqrt{2K}} $ and thereby establish the optimal minimax lower bound in one dimension. To establish the dimension-independent upper bound, we next show that a mini-batching algorithm provides an $ O(\frac{T}{\sqrt{K}}) $ upper bound, and therefore conclude that the minimax regret of switching-constrained OCO is $ Θ(\frac{T}{\sqrt{K}}) $ for any $K$. This is in sharp contrast to its discrete counterpart, the switching-constrained prediction-from-experts problem, which exhibits a phase transition in minimax regret between the low-switching and high-switching regimes.

cs.LG↗

arXiv2020

$k$-Uniform states and quantum information masking

A pure state of $N$ parties with local dimension $d$ is called a $k$-uniform state if all the reductions to $k$ parties are maximally mixed. Based on the connections among $k$-uniform states, orthogonal arrays and linear codes, we give general constructions for $k$-uniform states. We show that when $d\geq 4k-2$ (resp. $d\geq 2k-1$) is a prime power, there exists a $k$-uniform state for any $N\geq 2k$ (resp. $2k\leq N\leq d+1$). Specially, we give the existence of $4,5$-uniform states for almost every $N$-qudits. Further, we generalize the concept of quantum information masking in bipartite systems given by [Modi \emph{et al.} {Phys. Rev. Lett. \textbf{120}, 230501 (2018)}] to $k$-uniform quantum information masking in multipartite systems, and we show that $k$-uniform states and quantum error-correcting codes can be used for $k$-uniform quantum information masking.

quant-ph↗

arXiv2020

Region Comparison Network for Interpretable Few-shot Image Classification

While deep learning has been successfully applied to many real-world computer vision tasks, training robust classifiers usually requires a large amount of well-labeled data. However, the annotation is often expensive and time-consuming. Few-shot image classification has thus been proposed to effectively use only a limited number of labeled examples to train models for new classes. Recent works based on transferable metric learning methods have achieved promising classification performance through learning the similarity between the features of samples from the query and support sets. However, rare of them explicitly considers the model interpretability, which can actually be revealed during the training phase. For that, in this work, we propose a metric learning based method named Region Comparison Network (RCN), which is able to reveal how few-shot learning works as in a neural network as well as to find out specific regions that are related to each other in images coming from the query and support sets. Moreover, we also present a visualization strategy named Region Activation Mapping (RAM) to intuitively explain what our method has learned by visualizing intermediate variables in our network. We also present a new way to generalize the interpretability from the level of tasks to categories, which can also be viewed as a method to find the prototypical parts for supporting the final decision of our RCN. Extensive experiments on four benchmark datasets clearly show the effectiveness of our method over existing baselines.

cs.CV↗

arXiv2020

More Data Can Expand the Generalization Gap Between Adversarially Robust and Standard Models

Despite remarkable success in practice, modern machine learning models have been found to be susceptible to adversarial attacks that make human-imperceptible perturbations to the data, but result in serious and potentially dangerous prediction errors. To address this issue, practitioners often use adversarial training to learn models that are robust against such attacks at the cost of higher generalization error on unperturbed test sets. The conventional wisdom is that more training data should shrink the gap between the generalization error of adversarially-trained models and standard models. However, we study the training of robust classifiers for both Gaussian and Bernoulli models under $\ell_\infty$ attacks, and we prove that more data may actually increase this gap. Furthermore, our theoretical results identify if and when additional data will finally begin to shrink the gap. Lastly, we experimentally demonstrate that our results also hold for linear regression models, which may indicate that this phenomenon occurs more broadly.

cs.LG↗

arXiv2020

An Extention of Entanglement Measures for Pure States

To quantify the entanglement is one of the most important topics in quantum entanglement theory. In [arXiv: 2006.12408], the authors proposed a method to build a measure from the orginal domain to a larger one. Here we apply that method to build an entanglement measure from measures for pure states. First, we present conditions when the entanglement measure is an entanglement monotone and convex, we also present an interpretation of the smoothed one-shot entanglement cost under the method here. At last, we present a difference between the local operation and classical communication (LOCC) and the separability-preserving (SEPP) operations, then we present the entanglement measures built from the geometric entanglement measure for pure states by the convex roof extended method and the method here are equal, at last, we present the relationship between the concurrence and the entanglement measure built from concurrence for pure states by the method here on 2 \otimes 2 systems. We also present the measure is monogamous for 2\otimes 2 \otimes d system.

quant-ph↗

arXiv2020

Strong quantum nonlocality with entanglement

Strong quantum nonlocality was introduced recently as a stronger manifestation of nonlocality in multipartite systems through the notion of local irreducibility in all bipartitions. Known existence results for sets of strongly nonlocal orthogonal states are limited to product states. In this paper, based on the Rubik's cube, we give the first construction of such sets consisting of entangled states in $d\otimes d\otimes d$ for all $d\geq 3$. Consequently, we answer an open problem given by Halder \emph{et al.} [Phys. Rev. Lett. \textbf{122}, 040403 (2019)], that is, orthogonal entangled bases that are strongly nonlocal do exist. Furthermore, we propose two entanglement-assisted protocols for local discrimination of our results. Each protocol consumes less entanglement resource than the teleportation-based protocol averagely. Our results exhibit the phenomenon of strong quantum nonlocality with entanglement.

quant-ph↗

arXiv2020

Quasibound states in the continuum in terahertz free-standing metal complementary periodic cross-shaped resonators

We numerically and experimentally achieve quasi-bound states in the continuums (BICs) with high-Q factors in the free-standing metal complementary periodic cross-shaped resonators (CPCRs) at terahertz (THz) frequencies. Such induced quasi-BICs arises from the breaking of the mirror symmetry of CPCRs. By properly tuning the asymmetric factor, the measured Q factor of quasi-BIC can reach 102, which is lower than the simulated Q factor of 166 due to the limited system resolutions. We also simulate the electric field magnitude and vector distributions at the quasi-BICs, where the out-phase alignment between the electric dipoles is found. The sharp quasi-BICs realized in this thin free-standing metal structure may immediately boost the performance of filters and sensors in terahertz wave manipulation or biomolecular sensing.

physics.optics↗

arXiv2020

Joint Multi-User DNN Partitioning and Computational Resource Allocation for Collaborative Edge Intelligence

Mobile Edge Computing (MEC) has emerged as a promising supporting architecture providing a variety of resources to the network edge, thus acting as an enabler for edge intelligence services empowering massive mobile and Internet of Things (IoT) devices with AI capability. With the assistance of edge servers, user equipments (UEs) are able to run deep neural network (DNN) based AI applications, which are generally resource-hungry and compute-intensive, such that an individual UE can hardly afford by itself in real time. However the resources in each individual edge server are typically limited. Therefore, any resource optimization involving edge servers is by nature a resource-constrained optimization problem and needs to be tackled in such realistic context. Motivated by this observation, we investigate the optimization problem of DNN partitioning (an emerging DNN offloading scheme) in a realistic multi-user resource-constrained condition that rarely considered in previous works. Despite the extremely large solution space, we reveal several properties of this specific optimization problem of joint multi-UE DNN partitioning and computational resource allocation. We propose an algorithm called Iterative Alternating Optimization (IAO) that can achieve the optimal solution in polynomial time. In addition, we present rigorous theoretic analysis of our algorithm in terms of time complexity and performance under realistic estimation error. Moreover, we build a prototype that implements our framework and conduct extensive experiments using realistic DNN models, whose results demonstrate its effectiveness and efficiency.

cs.DC↗

arXiv2020

Meta Learning in the Continuous Time Limit

In this paper, we establish the ordinary differential equation (ODE) that underlies the training dynamics of Model-Agnostic Meta-Learning (MAML). Our continuous-time limit view of the process eliminates the influence of the manually chosen step size of gradient descent and includes the existing gradient descent training algorithm as a special case that results from a specific discretization. We show that the MAML ODE enjoys a linear convergence rate to an approximate stationary point of the MAML loss function for strongly convex task losses, even when the corresponding MAML loss is non-convex. Moreover, through the analysis of the MAML ODE, we propose a new BI-MAML training algorithm that significantly reduces the computational burden associated with existing MAML training methods. To complement our theoretical findings, we perform empirical experiments to showcase the superiority of our proposed methods with respect to the existing work.

stat.ML↗

arXiv2020

Computational Complexity Characterization of Protecting Elections from Bribery

The bribery problem in election has received considerable attention in the literature, upon which various algorithmic and complexity results have been obtained. It is thus natural to ask whether we can protect an election from potential bribery. We assume that the protector can protect a voter with some cost (e.g., by isolating the voter from potential bribers). A protected voter cannot be bribed. Under this setting, we consider the following bi-level decision problem: Is it possible for the protector to protect a proper subset of voters such that no briber with a fixed budget on bribery can alter the election result? The goal of this paper is to give a full picture on the complexity of protection problems. We give an extensive study on the protection problem and provide algorithmic and complexity results. Comparing our results with that on the bribery problems, we observe that the protection problem is in general significantly harder. Indeed, it becomes $\sum_{p}^2$-complete even for very restricted special cases, while most bribery problems lie in NP. However, it is not necessarily the case that the protection problem is always harder. Some of the protection problems can still be solved in polynomial time, while some of them remain as hard as the bribery problem under the same setting.

cs.CC↗