Searcharxiv⌕ Search

arXiv subjects

Cheng Zhang

Publications and source records attributed to Cheng Zhang.

At least 415 records · Page 23Linked to original sources
arXiv2020

How Do Fair Decisions Fare in Long-term Qualification?

Although many fairness criteria have been proposed for decision making, their long-term impact on the well-being of a population remains unclear. In this work, we study the dynamics of population qualification and algorithmic decisions under a partially observed Markov decision problem setting. By characterizing the equilibrium of such dynamics, we analyze the long-term impact of static fairness constraints on the equality and improvement of group well-being. Our results show that static fairness constraints can either promote equality or exacerbate disparity depending on the driving factor of qualification transitions and the effect of sensitive attributes on feature distributions. We also consider possible interventions that can effectively improve group qualification or promote equality of group qualification. Our theoretical results and experiments on static real-world datasets with simulated dynamics show that our framework can be used to facilitate social science studies.

cs.LG↗
arXiv2020

Progressive Defense Against Adversarial Attacks for Deep Learning as a Service in Internet of Things

Nowadays, Deep Learning as a service can be deployed in Internet of Things (IoT) to provide smart services and sensor data processing. However, recent research has revealed that some Deep Neural Networks (DNN) can be easily misled by adding relatively small but adversarial perturbations to the input (e.g., pixel mutation in input images). One challenge in defending DNN against these attacks is to efficiently identifying and filtering out the adversarial pixels. The state-of-the-art defense strategies with good robustness often require additional model training for specific attacks. To reduce the computational cost without loss of generality, we present a defense strategy called a progressive defense against adversarial attacks (PDAAA) for efficiently and effectively filtering out the adversarial pixel mutations, which could mislead the neural network towards erroneous outputs, without a-priori knowledge about the attack type. We evaluated our progressive defense strategy against various attack methods on two well-known datasets. The result shows it outperforms the state-of-the-art while reducing the cost of model training by 50% on average.

cs.CR↗
arXiv2020

Weakly-supervised Fine-grained Event Recognition on Social Media Texts for Disaster Management

People increasingly use social media to report emergencies, seek help or share information during disasters, which makes social networks an important tool for disaster management. To meet these time-critical needs, we present a weakly supervised approach for rapidly building high-quality classifiers that label each individual Twitter message with fine-grained event categories. Most importantly, we propose a novel method to create high-quality labeled data in a timely manner that automatically clusters tweets containing an event keyword and asks a domain expert to disambiguate event word senses and label clusters quickly. In addition, to process extremely noisy and often rather short user-generated messages, we enrich tweet representations using preceding context tweets and reply tweets in building event recognition classifiers. The evaluation on two hurricanes, Harvey and Florence, shows that using only 1-2 person-hours of human supervision, the rapidly trained weakly supervised classifiers outperform supervised classifiers trained using more than ten thousand annotated tweets created in over 50 person-hours.

cs.CL↗
arXiv2020

Spatio-Temporal Hierarchical Adaptive Dispatching for Ridesharing Systems

Nowadays, ridesharing has become one of the most popular services offered by online ride-hailing platforms (e.g., Uber and Didi Chuxing). Existing ridesharing platforms adopt the strategy that dispatches orders over the entire city at a uniform time interval. However, the uneven spatio-temporal order distributions in real-world ridesharing systems indicate that such an approach is suboptimal in practice. Thus, in this paper, we exploit adaptive dispatching intervals to boost the platform's profit under a guarantee of the maximum passenger waiting time. Specifically, we propose a hierarchical approach, which generates clusters of geographical areas suitable to share the same dispatching intervals, and then makes online decisions of selecting the appropriate time instances for order dispatch within each spatial cluster. Technically, we prove the impossibility of designing constant-competitive-ratio algorithms for the online adaptive interval problem, and propose online algorithms under partial or even zero future order knowledge that significantly improve the platform's profit over existing approaches. We conduct extensive experiments with a large-scale ridesharing order dataset, which contains all of the over 3.5 million ridesharing orders in Beijing, China, received by Didi Chuxing from October 1st to October 31st, 2018. The experimental results demonstrate that our proposed algorithms outperform existing approaches.

eess.SP↗
arXiv2020

Integrable boundary conditions for quad equations, open boundary reductions and integrable mappings

In the context of integrable partial difference equations on quad-graphs, we introduce the notion of open boundary reductions as a new means to construct discrete integrable mappings and their invariants. This represents an alternative to the well-known periodic reductions. The construction deals with well-posed initial value problems for quad equations on quad-graphs restricted to a strip. It relies on the so-called double-row monodromy matrix and gives rise to integrable mappings. To obtain the double-row monodromy matrix, we use the notion of boundary matrix and discrete boundary zero curvature condition, themselves related to the boundary consistency condition, which complements the well-known $3$D consistency condition for integrable quad equations and gives an integrability criterion for boundary equations. This relation is made precise in this paper. Our focus is on quad equations from the Adler-Bobenko-Suris (ABS) classification and their discrete integrable boundary equations. Taking as our prime example a regular $\mathbb{Z}^2$-lattice with two parallel boundaries, we provide an explicit construction of the maps obtained by open boundary reductions, the boundary matrices, as well as the invariants extracted from the double-row monodromy matrix. Examples of integrable maps are considered for the H1 and Q1($δ=0$) equations and an interesting example of a non-QRT map of the plane is presented. Examples of well-posed quad-graph systems on a strip beyond the $\mathbb{Z}^2$-lattice are also given.

math-ph↗
arXiv2020

Thermodynamics of Chiral Fermion System in a Uniform Magnetic Field

We construct the grand partition function of the system of chiral fermions in a uniform magnetic field from Landau levels, through which all thermodynamic quantities can be obtained. Taking use of Abel-Plana formula, these thermodynamic quantities can be expanded as series with respect to a dimensionless variable $b=2eB/T^{2}$. We find that the series expansions of energy density, pressure, magnetization intensity and magnetic susceptibility contain a singular term with $\ln b^{2}$, while particle number density, entropy density and heat capacity are power series of $b^{2}$. The asymptotic behaviors of these thermodynamic quantities in extreme conditions are also discussed.

hep-th↗
arXiv2020

On the identifiability of interaction functions in systems of interacting particles

We address a fundamental issue in the nonparametric inference for systems of interacting particles: the identifiability of the interaction functions. We prove that the interaction functions are identifiable for a class of first-order stochastic systems, including linear systems with general initial laws and nonlinear systems with stationary distributions. We show that a coercivity condition is sufficient for identifiability and becomes necessary when the number of particles approaches infinity. The coercivity is equivalent to the strict positivity of related integral operators, which we prove by showing that their integral kernels are strictly positive definite by using Müntz type theorems.

math.ST↗
arXiv2020

The $\cos 2ϕ$ azimuthal asymmetry in $ρ^0$ meson production in ultraperipheral heavy ion collisions

We present a detailed study of vector meson photoproduction in ultraperipheral heavy ion collisions (UPCs). Using the dipole model, we develop a framework for the joint impact parameter and transverse momentum dependent cross sections. We compute the unpolarized cross section and $\cos 2ϕ$ azimuthal angular correlation for $ρ^0$ photoproduction with $ϕ$ defined as the angle between the $ρ^0$'s transverse momentum and its decay product pion meson's transverse momentum. Our result on unpolarized coherent differential cross section gives excellent description to the STAR experimental data. A first comparison between theoretical calculation and experimental measurement on the $\cos 2ϕ$ azimuthal asymmetry, which results from the linearly polarized photons, is performed and reasonable agreement is reached. We find out the characteristic diffractive patterns at both RHIC and LHC energies and predict the impact parameter dependent $\cos 2ϕ$ azimuthal asymmetries for $ρ^0$ photoproduction by considering UPCs and peripheral collisions. The future experimental measurements at RHIC and LHC relevant to our calculations will provide a tool to rigorously investigate the coherent and incoherent production of vector meson in UPCs, as well as to probe the nuclear structure in heavy ion collisions.

hep-ph↗
arXiv2020

Thoracic Disease Identification and Localization using Distance Learning and Region Verification

The identification and localization of diseases in medical images using deep learning models have recently attracted significant interest. Existing methods only consider training the networks with each image independently and most leverage an activation map for disease localization. In this paper, we propose an alternative approach that learns discriminative features among triplets of images and cyclically trains on region features to verify whether attentive regions contain information indicative of a disease. Concretely, we adapt a distance learning framework for multi-label disease classification to differentiate subtle disease features. Additionally, we feed back the features of the predicted class-specific regions to a separate classifier during training to better verify the localized diseases. Our model can achieve state-of-the-art classification performance on the challenging ChestX-ray14 dataset, and our ablation studies indicate that both distance learning and region verification contribute to overall classification performance. Moreover, the distance learning and region verification modules can capture essential information for better localization than baseline models without these modules.

cs.CV↗
arXiv2020

Intervention Generative Adversarial Networks

In this paper we propose a novel approach for stabilizing the training process of Generative Adversarial Networks as well as alleviating the mode collapse problem. The main idea is to introduce a regularization term that we call intervention loss into the objective. We refer to the resulting generative model as Intervention Generative Adversarial Networks (IVGAN). By perturbing the latent representations of real images obtained from an auxiliary encoder network with Gaussian invariant interventions and penalizing the dissimilarity of the distributions of the resulting generated images, the intervention loss provides more informative gradient for the generator, significantly improving GAN's training stability. We demonstrate the effectiveness and efficiency of our methods via solid theoretical analysis and thorough evaluation on standard real-world datasets as well as the stacked MNIST dataset.

stat.ML↗
arXiv2020

Hide-and-Seek Privacy Challenge

The clinical time-series setting poses a unique combination of challenges to data modeling and sharing. Due to the high dimensionality of clinical time series, adequate de-identification to preserve privacy while retaining data utility is difficult to achieve using common de-identification techniques. An innovative approach to this problem is synthetic data generation. From a technical perspective, a good generative model for time-series data should preserve temporal dynamics, in the sense that new sequences respect the original relationships between high-dimensional variables across time. From the privacy perspective, the model should prevent patient re-identification by limiting vulnerability to membership inference attacks. The NeurIPS 2020 Hide-and-Seek Privacy Challenge is a novel two-tracked competition to simultaneously accelerate progress in tackling both problems. In our head-to-head format, participants in the synthetic data generation track (i.e. "hiders") and the patient re-identification track (i.e. "seekers") are directly pitted against each other by way of a new, high-quality intensive care time-series dataset: the AmsterdamUMCdb dataset. Ultimately, we seek to advance generative techniques for dense and high-dimensional temporal data streams that are (1) clinically meaningful in terms of fidelity and predictivity, as well as (2) capable of minimizing membership privacy risks in terms of the concrete notion of patient re-identification.

cs.LG↗
arXiv2020

DRIFT: Deep Reinforcement Learning for Functional Software Testing

Efficient software testing is essential for productive software development and reliable user experiences. As human testing is inefficient and expensive, automated software testing is needed. In this work, we propose a Reinforcement Learning (RL) framework for functional software testing named DRIFT. DRIFT operates on the symbolic representation of the user interface. It uses Q-learning through Batch-RL and models the state-action value function with a Graph Neural Network. We apply DRIFT to testing the Windows 10 operating system and show that DRIFT can robustly trigger the desired software functionality in a fully automated manner. Our experiments test the ability to perform single and combined tasks across different applications, demonstrating that our framework can efficiently test software with a large range of testing objectives.

cs.LG↗
arXiv2020

Causal Discovery in the Presence of Missing Data

Missing data are ubiquitous in many domains including healthcare. When these data entries are not missing completely at random, the (conditional) independence relations in the observed data may be different from those in the complete data generated by the underlying causal process. Consequently, simply applying existing causal discovery methods to the observed data may lead to wrong conclusions. In this paper, we aim at developing a causal discovery method to recover the underlying causal structure from observed data that follow different missingness mechanisms, including missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). With missingness mechanisms represented by missingness graphs, we analyse conditions under which additional correction is needed to derive conditional independence/dependence relations in the complete data. Based on our analysis, we propose the Missing Value PC (MVPC) algorithm for both continuous and binary variables, which extends the PC algorithm to incorporate additional corrections. Our proposed MVPC is shown in theory to give asymptotically correct results even on data that are MAR or MNAR. Experimental results on synthetic data show that the proposed algorithm is able to find correct causal relations even in the general case of MNAR. Moreover, we create a neuropathic pain diagnostic simulator for evaluating causal discovery methods. Evaluated on such simulated neuropathic pain diagnosis records and the other two real world applications, MVPC outperforms the other benchmark methods.

cs.LG↗
arXiv2020

ACOUSTIC-TURF: Acoustic-based Privacy-Preserving COVID-19 Contact Tracing

In this paper, we propose a new privacy-preserving, automated contact tracing system, ACOUSTIC-TURF, to fight COVID-19 using acoustic signals sent from ubiquitous mobile devices. At a high level, ACOUSTIC-TURF adaptively broadcasts inaudible ultrasonic signals with randomly generated IDs in the vicinity. Simultaneously, the system receives other ultrasonic signals sent from nearby (e.g., 6 feet) users. In such a system, individual user IDs are not disclosed to others and the system can accurately detect encounters in physical proximity with 6-foot granularity. We have implemented a prototype of ACOUSTIC-TURF on Android and evaluated its performance in terms of acoustic-signal-based encounter detection accuracy and power consumption at different ranges and under various occlusion scenarios. Experimental results show that ACOUSTIC-TURF can detect multiple contacts within a 6-foot range for mobile phones placed in pockets and outside pockets. Furthermore, our acoustic-signal-based system achieves greater precision than wireless-signal-based approaches when contact tracing is performed through walls. ACOUSTIC-TURF correctly determines that people on opposite sides of a wall are not in contact with one another, whereas the Bluetooth-based approaches detect nonexistent contacts among them.

cs.CR↗
arXiv2020

VAEM: a Deep Generative Model for Heterogeneous Mixed Type Data

Deep generative models often perform poorly in real-world applications due to the heterogeneity of natural data sets. Heterogeneity arises from data containing different types of features (categorical, ordinal, continuous, etc.) and features of the same type having different marginal distributions. We propose an extension of variational autoencoders (VAEs) called VAEM to handle such heterogeneous data. VAEM is a deep generative model that is trained in a two stage manner such that the first stage provides a more uniform representation of the data to the second stage, thereby sidestepping the problems caused by heterogeneous data. We provide extensions of VAEM to handle partially observed data, and demonstrate its performance in data generation, missing data prediction and sequential feature selection tasks. Our results show that VAEM broadens the range of real-world applications where deep generative models can be successfully deployed.

cs.LG↗
arXiv2020

Sharp endpoint estimates for eigenfunctions restricted to submanifolds of codimension 2

Burq-Gérard-Tzvetkov and Hu established $L^p$ estimates ($2\le p\le \infty$) for the restriction of eigenfunctions to submanifolds. The estimates are sharp, except for the log loss at the endpoint $L^2$ estimates for submanifolds of codimension 2. It has long been believed that the log loss at the endpoint can be removed in general, while the problem is still open. So this paper is devoted to the study of sharp endpoint restriction estimates for eigenfunctions in this case. Chen and Sogge removed the log loss for the geodesics on 3-dimensional manifolds. In this paper, we generalize their result to higher dimensions and prove that the log loss can be removed for totally geodesic submanifolds of codimension 2. Moreover, on 3-dimensional manifolds, we can remove the log loss for curves with nonvanishing geodesic curvatures, and more general finite type curves. The problem in 3D is essentially related to Hilbert transforms along curves in the plane and a class of singular oscillatory integrals studied by Phong-Stein, Ricci-Stein, Pan, Seeger, Carbery-Pérez.

math.AP↗
arXiv2020

Estimating a fluctuating magnetic field with a continuously monitored atomic ensemble

We study the problem of estimating a time dependent magnetic field by continuous optical probing of an atomic ensemble. The magnetic field is assumed to follow a stochastic Ornstein-Uhlenbeck process and it induces Larmor precession of the atomic ground state spin, which is read out by the Faraday polarization rotation of a laser field probe. The interactions and the measurement scheme are compatible with a hybrid quantum-classical Gaussian description of the unknown magnetic field, and the atomic and field variables. This casts the joint conditional quantum dynamics and classical parameter estimation problem in the form of update formulas for the first and second moments of the classical and quantum degrees of freedom. Our hybrid quantum-classical theory is equivalent with the classical theory of Kalman filtering and with the quantum theory of Gaussian states. By reference to the classical theory of smoothing and with the quantum theory of past quantum states, we show how optical probing after time $t$ improves our estimate of the value of the magnetic field at time $t$, and we present numerical simulations that analyze and explain the improvement over the conventional filtering approach.

quant-ph↗