arXiv subjects

Ting Wang

Publications and source records attributed to Ting Wang.

At least 217 records · Page 12Linked to original sources

arXiv2021

Layer-splitting methods for time-dependent Schrödinger equations of incommensurate systems

This work considers numerical methods for the time-dependent Schrödinger equation of incommensurate systems. By using a plane wave method for spatial discretization, the incommensurate problem is lifted to a higher dimension that results in semidiscrete differential equations with extremely demanding computational cost. We propose several fully discrete time stepping schemes based on the idea of "layer-splitting", which decompose the semidiscrete problem into sub-problems that each corresponds to one of the periodic layers. Then these schemes handle only some periodic systems in the original lower dimension at each time step, which reduces the computational cost significantly and is natural to involve stochastic methods and parallel computing. Both theoretical analysis and numerical experiments are provided to support the reliability and efficiency of the algorithms.

physics.comp-ph↗

arXiv2021

Trojaning Language Models for Fun and Profit

Recent years have witnessed the emergence of a new paradigm of building natural language processing (NLP) systems: general-purpose, pre-trained language models (LMs) are composed with simple downstream models and fine-tuned for a variety of NLP tasks. This paradigm shift significantly simplifies the system development cycles. However, as many LMs are provided by untrusted third parties, their lack of standardization or regulation entails profound security implications, which are largely unexplored. To bridge this gap, this work studies the security threats posed by malicious LMs to NLP systems. Specifically, we present TROJAN-LM, a new class of trojaning attacks in which maliciously crafted LMs trigger host NLP systems to malfunction in a highly predictable manner. By empirically studying three state-of-the-art LMs (BERT, GPT-2, XLNet) in a range of security-critical NLP tasks (toxic comment detection, question answering, text completion) as well as user studies on crowdsourcing platforms, we demonstrate that TROJAN-LM possesses the following properties: (i) flexibility - the adversary is able to flexibly dene logical combinations (e.g., 'and', 'or', 'xor') of arbitrary words as triggers, (ii) efficacy - the host systems misbehave as desired by the adversary with high probability when trigger-embedded inputs are present, (iii) specificity - the trojan LMs function indistinguishably from their benign counterparts on clean inputs, and (iv) fluency - the trigger-embedded inputs appear as fluent natural language and highly relevant to their surrounding contexts. We provide analytical justification for the practicality of TROJAN-LM, and further discuss potential countermeasures and their challenges, which lead to several promising research directions.

cs.CR↗

arXiv2021

Position-dependent chiral coupling between single quantum dots and cross waveguides

Chiral light-matter interaction between photonic nanostructures with quantum emitters shows great potential to implement spin-photon interfaces for quantum information processing. Position-dependent spin momentum locking of the quantum emitter is important for these chiral coupled nanostructures. Here, we report the position-dependent chiral coupling between quantum dots (QDs) and cross waveguides both numerically and experimentally. Four quantum dots distributed at different positions in the cross section are selected to characterize the chiral properties of the device. Directional emission is achieved in a single waveguide as well as in both two waveguides simultaneously. In addition, the QD position can be determined with the chiral contrasts from four outputs. Therefore, the cross waveguide can function as a one-way unidirectional waveguide and a circularly polarized beam splitter by placing the QD in a rational position, which has potential applications in spin-to-path encoding for complex quantum optical networks at the single-photon level.

cond-mat.mes-hall↗

arXiv2021

UAV-Assisted Over-the-Air Computation

Over-the-air computation (AirComp) provides a promising way to support ultrafast aggregation of distributed data. However, its performance cannot be guaranteed in long-distance transmission due to the distortion induced by the channel fading and noise. To unleash the full potential of AirComp, this paper proposes to use a low-cost unmanned aerial vehicle (UAV) acting as a mobile base station to assist AirComp systems. Specifically, due to its controllable high-mobility and high-altitude, the UAV can move sufficiently close to the sensors to enable line-of-sight transmission and adaptively adjust all the links' distances, thereby enhancing the signal magnitude alignment and noise suppression. Our goal is to minimize the time-averaging mean-square error for AirComp by jointly optimizing the UAV trajectory, the scaling factor at the UAV, and the transmit power at the sensors, under constraints on the UAV's predetermined locations and flying speed, sensors' average and peak power limits. However, due to the highly coupled optimization variables and time-dependent constraints, the resulting problem is non-convex and challenging. We thus propose an efficient iterative algorithm by applying the block coordinate descent and successive convex optimization techniques. Simulation results verify the convergence of the proposed algorithm and demonstrate the performance gains and robustness of the proposed design compared with benchmarks.

eess.SP↗

arXiv2021

i-Algebra: Towards Interactive Interpretability of Deep Neural Networks

Providing explanations for deep neural networks (DNNs) is essential for their use in domains wherein the interpretability of decisions is a critical prerequisite. Despite the plethora of work on interpreting DNNs, most existing solutions offer interpretability in an ad hoc, one-shot, and static manner, without accounting for the perception, understanding, or response of end-users, resulting in their poor usability in practice. In this paper, we argue that DNN interpretability should be implemented as the interactions between users and models. We present i-Algebra, a first-of-its-kind interactive framework for interpreting DNNs. At its core is a library of atomic, composable operators, which explain model behaviors at varying input granularity, during different inference stages, and from distinct interpretation perspectives. Leveraging a declarative query language, users are enabled to build various analysis tools (e.g., "drill-down", "comparative", "what-if" analysis) via flexibly composing such operators. We prototype i-Algebra and conduct user studies in a set of representative analysis tasks, including inspecting adversarial inputs, resolving model inconsistency, and cleansing contaminated data, all demonstrating its promising usability.

cs.LG↗

arXiv2021

Visual Perception Generalization for Vision-and-Language Navigation via Meta-Learning

Vision-and-language navigation (VLN) is a challenging task that requires an agent to navigate in real-world environments by understanding natural language instructions and visual information received in real-time. Prior works have implemented VLN tasks on continuous environments or physical robots, all of which use a fixed camera configuration due to the limitations of datasets, such as 1.5 meters height, 90 degrees horizontal field of view (HFOV), etc. However, real-life robots with different purposes have multiple camera configurations, and the huge gap in visual information makes it difficult to directly transfer the learned navigation model between various robots. In this paper, we propose a visual perception generalization strategy based on meta-learning, which enables the agent to fast adapt to a new camera configuration with a few shots. In the training phase, we first locate the generalization problem to the visual perception module, and then compare two meta-learning algorithms for better generalization in seen and unseen environments. One of them uses the Model-Agnostic Meta-Learning (MAML) algorithm that requires a few shot adaptation, and the other refers to a metric-based meta-learning method with a feature-wise affine transformation layer. The experiment results show that our strategy successfully adapts the learned navigation model to a new camera configuration, and the two algorithms show their advantages in seen and unseen environments respectively.

cs.RO↗

arXiv2020

A Tale of Evil Twins: Adversarial Inputs versus Poisoned Models

Despite their tremendous success in a range of domains, deep learning systems are inherently susceptible to two types of manipulations: adversarial inputs -- maliciously crafted samples that deceive target deep neural network (DNN) models, and poisoned models -- adversely forged DNNs that misbehave on pre-defined inputs. While prior work has intensively studied the two attack vectors in parallel, there is still a lack of understanding about their fundamental connections: what are the dynamic interactions between the two attack vectors? what are the implications of such interactions for optimizing existing attacks? what are the potential countermeasures against the enhanced attacks? Answering these key questions is crucial for assessing and mitigating the holistic vulnerabilities of DNNs deployed in realistic settings. Here we take a solid step towards this goal by conducting the first systematic study of the two attack vectors within a unified framework. Specifically, (i) we develop a new attack model that jointly optimizes adversarial inputs and poisoned models; (ii) with both analytical and empirical evidence, we reveal that there exist intriguing "mutual reinforcement" effects between the two attack vectors -- leveraging one vector significantly amplifies the effectiveness of the other; (iii) we demonstrate that such effects enable a large design spectrum for the adversary to enhance the existing attacks that exploit both vectors (e.g., backdoor attacks), such as maximizing the attack evasiveness with respect to various detection methods; (iv) finally, we discuss potential countermeasures against such optimized attacks and their technical challenges, pointing to several promising research directions.

cs.LG↗

arXiv2020

On the Validity of the Girsanov Transformation Method for Sensitivity Analysis of Stochastic Chemical Reaction Networks

We investigate the validity of the Girsanov Transformation (GT) method for parametric sensitivity analysis of stochastic models of chemical reaction networks. The validity depends on the likelihood ratio process being a martingale and the commutation of a certain derivative with expectation. We derive some exponential integrability conditions which imply both these requirements. We provide further conditions in terms of a reaction network that imply these exponential integrability conditions.

math.PR↗

arXiv2020

On Efficient and Scalable Time-Continuous Spatial Crowdsourcing -- Full Version

The proliferation of advanced mobile terminals opened up a new crowdsourcing avenue, spatial crowdsourcing, to utilize the crowd potential to perform real-world tasks. In this work, we study a new type of spatial crowdsourcing, called time-continuous spatial crowdsourcing (TCSC in short). It supports broad applications for long-term continuous spatial data acquisition, ranging from environmental monitoring to traffic surveillance in citizen science and crowdsourcing projects. However, due to limited budgets and limited availability of workers in practice, the data collected is often incomplete, incurring data deficiency problem. To tackle that, in this work, we first propose an entropy-based quality metric, which captures the joint effects of incompletion in data acquisition and the imprecision in data interpolation. Based on that, we investigate quality-aware task assignment methods for both single- and multi-task scenarios. We show the NP-hardness of the single-task case, and design polynomial-time algorithms with guaranteed approximation ratios. We study novel indexing and pruning techniques for further enhancing the performance in practice. Then, we extend the solution to multi-task scenarios and devise a parallel framework for speeding up the process of optimization. We conduct extensive experiments on both real and synthetic datasets to show the effectiveness of our proposals.

cs.DB↗

arXiv2020

UNIFUZZ: A Holistic and Pragmatic Metrics-Driven Platform for Evaluating Fuzzers

A flurry of fuzzing tools (fuzzers) have been proposed in the literature, aiming at detecting software vulnerabilities effectively and efficiently. To date, it is however still challenging to compare fuzzers due to the inconsistency of the benchmarks, performance metrics, and/or environments for evaluation, which buries the useful insights and thus impedes the discovery of promising fuzzing primitives. In this paper, we design and develop UNIFUZZ, an open-source and metrics-driven platform for assessing fuzzers in a comprehensive and quantitative manner. Specifically, UNIFUZZ to date has incorporated 35 usable fuzzers, a benchmark of 20 real-world programs, and six categories of performance metrics. We first systematically study the usability of existing fuzzers, find and fix a number of flaws, and integrate them into UNIFUZZ. Based on the study, we propose a collection of pragmatic performance metrics to evaluate fuzzers from six complementary perspectives. Using UNIFUZZ, we conduct in-depth evaluations of several prominent fuzzers including AFL [1], AFLFast [2], Angora [3], Honggfuzz [4], MOPT [5], QSYM [6], T-Fuzz [7] and VUzzer64 [8]. We find that none of them outperforms the others across all the target programs, and that using a single metric to assess the performance of a fuzzer may lead to unilateral conclusions, which demonstrates the significance of comprehensive metrics. Moreover, we identify and investigate previously overlooked factors that may significantly affect a fuzzer's performance, including instrumentation methods and crash analysis tools. Our empirical results show that they are critical to the evaluation of a fuzzer. We hope that our findings can shed light on reliable fuzzing evaluation, so that we can discover promising fuzzing primitives to effectively facilitate fuzzer designs in the future.

cs.CR↗

arXiv2020

Dipole coupling of a tunable hole double quantum dot in germanium hut wire to a microwave resonator

The germanium (Ge) hut wire system has strong spin-orbit coupling, a long coherence time due to a very large heavy-light hole splitting, and the advantage of site-controlled large-scale hut wire positioning. These properties make the Ge hut wire a promising candidate for the realization of strong coupling of spin to superconducting resonators and scalability for multiple qubit coupling. We have coupled a reflection line resonator to a hole double quantum dot (DQD) formed in Ge hut wire. The amplitude and phase responses of the microwave resonator revealed that the charge stability diagrams of the DQD are in good agreement with those obtained from transport measurements. The DQD interdot tunneling rate is shown to be tunable from 6.2 GHz to 8.5 GHz, which demonstrates the ability to adjust the frequency detuning between the qubit and the resonator. Furthermore, we achieved a hole-resonator coupling strength of up to 15 MHz, with a charge qubit decoherence rate of 0.28 GHz. Meanwhile the hole spin-resonator coupling rate was estimated to be 3 MHz. These results suggest that holes of a DQD in a Ge hut wire are dipole coupled to microwave photons, potentially enabling tunable hole spin-photon interactions in Ge with an inherent spin-orbit coupling.

cond-mat.mes-hall↗

arXiv2020

PoisHygiene: Detecting and Mitigating Poisoning Attacks in Neural Networks

The black-box nature of deep neural networks (DNNs) facilitates attackers to manipulate the behavior of DNN through data poisoning. Being able to detect and mitigate poisoning attacks, typically categorized into backdoor and adversarial poisoning (AP), is critical in enabling safe adoption of DNNs in many application domains. Although recent works demonstrate encouraging results on detection of certain backdoor attacks, they exhibit inherent limitations which may significantly constrain the applicability. Indeed, no technique exists for detecting AP attacks, which represents a harder challenge given that such attacks exhibit no common and explicit rules while backdoor attacks do (i.e., embedding backdoor triggers into poisoned data). We believe the key to detect and mitigate AP attacks is the capability of observing and leveraging essential poisoning-induced properties within an infected DNN model. In this paper, we present PoisHygiene, the first effective and robust detection and mitigation framework against AP attacks. PoisHygiene is fundamentally motivated by Dr. Ernest Rutherford's story (i.e., the 1908 Nobel Prize winner), on observing the structure of atom through random electron sampling.

cs.CR↗

arXiv2020

AdvMind: Inferring Adversary Intent of Black-Box Attacks

Deep neural networks (DNNs) are inherently susceptible to adversarial attacks even under black-box settings, in which the adversary only has query access to the target models. In practice, while it may be possible to effectively detect such attacks (e.g., observing massive similar but non-identical queries), it is often challenging to exactly infer the adversary intent (e.g., the target class of the adversarial example the adversary attempts to craft) especially during early stages of the attacks, which is crucial for performing effective deterrence and remediation of the threats in many scenarios. In this paper, we present AdvMind, a new class of estimation models that infer the adversary intent of black-box adversarial attacks in a robust and prompt manner. Specifically, to achieve robust detection, AdvMind accounts for the adversary adaptiveness such that her attempt to conceal the target will significantly increase the attack cost (e.g., in terms of the number of queries); to achieve prompt detection, AdvMind proactively synthesizes plausible query results to solicit subsequent queries from the adversary that maximally expose her intent. Through extensive empirical evaluation on benchmark datasets and state-of-the-art black-box attacks, we demonstrate that on average AdvMind detects the adversary intent with over 75% accuracy after observing less than 3 query batches and meanwhile increases the cost of adaptive attacks by over 60%. We further discuss the possible synergy between AdvMind and other defense methods against black-box adversarial attacks, pointing to several promising research directions.

cs.LG↗

arXiv2020

Stochastic Gradient Descent for Semilinear Elliptic Equations with Uncertainties

Randomness is ubiquitous in modern engineering. The uncertainty is often modeled as random coefficients in the differential equations that describe the underlying physics. In this work, we describe a two-step framework for numerically solving semilinear elliptic partial differential equations with random coefficients: 1) reformulate the problem as a functional minimization problem based on the direct method of calculus of variation; 2) solve the minimization problem using the stochastic gradient descent method. We provide the convergence criterion for the resulted stochastic gradient descent algorithm and discuss some useful technique to overcome the issues of ill-conditioning and large variance. The accuracy and efficiency of the algorithm are demonstrated by numerical experiments.

math.NA↗

arXiv2020

Hole spin in tunable Ge hut wire double quantum dot

Holes in germanium (Ge) exhibit strong spin-orbit interaction, which can be exploited for fast and all-electrical manipulation of spin states. Here, we report transport experiments in a tunable Ge hut wire hole double quantum dot. We observe the signatures of Pauli spin blockade (PSB) with a large singlet-triplet energy splitting of ~1.1 meV and extract the g factor. By analyzing the the PSB leakage current, we obtain a spin-orbit length l_so of ~ 40-100 nm. Furthermore, we demonstrate the electric dipole spin resonance. These results lay a solid foundation for implementing high quality tunable hole spin-orbit qubits.

cond-mat.mes-hall↗

arXiv2020

Portably parallel construction of a CI wave function from a matrix-product state using the Charm++ framework

The constructions of configuration interaction (CI) expansions from a matrix-product state (MPS) involves numerous matrix operations and the skillful sampling of important configurations when in a huge Hilbert space. In this work, we present an efficient procedure for constructing CI expansions from MPS using the Charm++ parallel programming framework, upon which automatic load balancing and object migration facilities can be employed. This procedure was employed in the MPS-to-CI utility (Moritz et al., J. Chem. Phys. 2007, 126, 224109), sampling-reconstructed complete active space algorithm (SR-CAS, Boguslawski et al., J. Chem. Phys. 2011, 134, 224101) and entanglement-driven genetic algorithm (EDGA, Luo et al., J. Chem. Theory Comput. 2017, 13, 4699-4710). It enhances productivity and allows the sampling programs evolve to their population-expansion versions (e.g., EDGA with population expansion [PE-EDGA]). Examples of 1,2-dioxetanone and firefly dioxetanone anion (FDO-) molecules demonstrated that 1) the procedure could be flexibly employed among various multi-core architectures; 2) the parallel efficiencies could be persistently improved simply by increasing the proportion of asynchronous executions; 3) PE-EDGA could construct a CAS-type CI wave function from a huge Hilbert space, with 0.9952 CI completeness and 96.7% correlation energy via ~1.66x10^6 configurations (only 0.0000028% of the total configurations) of a bi-radical state of FDO- molecule using the full valence active space within a few hours.

physics.comp-ph↗

arXiv2020

A 1000-fold Acceleration of Hidden Markov Model Fitting using Graphical Processing Units, with application to Nonvolcanic Tremor Classification

Hidden Markov models (HMMs) are general purpose models for time-series data widely used across the sciences because of their flexibility and elegance. However fitting HMMs can often be computationally demanding and time consuming, particularly when the the number of hidden states is large or the Markov chain itself is long. Here we introduce a new Graphical Processing Unit (GPU) based algorithm designed to fit long chain HMMs, applying our approach to an HMM for nonvolcanic tremor events developed by Wang et al.(2018). Even on a modest GPU, our implementation resulted in a 1000-fold increase in speed over the standard single processor algorithm, allowing a full Bayesian inference of uncertainty related to model parameters. Similar improvements would be expected for HMM models given large number of observations and moderate state spaces (<80 states with current hardware). We discuss the model, general GPU architecture and algorithms and report performance of the method on a tremor dataset from the Shikoku region, Japan.

stat.CO↗

arXiv2020

Score-based Tests for Explaining Upper-Level Heterogeneity in Linear Mixed Models

Cross-level interactions among fixed effects in linear mixed models (also known as multilevel models) are often complicated by the variances stemming from random effects and residuals. When these variances change across clusters, tests of fixed effects (including cross-level interaction terms) are subject to inflated Type I or Type II error. While the impact of variance change/heterogeneity has been noticed in the literature, few methods have been proposed to detect this heterogeneity in a simple, systematic way. In addition, when heterogeneity among clusters is detected, researchers often wish to know which clusters' variances differed from the others. In this study, we utilize a recently-proposed family of score-based tests to distinguish between cross-level interactions and heterogeneity in variance components, also providing information about specific clusters that exhibit heterogeneity. These score-based tests only require estimation of the null model (when variance homogeneity is assumed to hold), and they have been previously applied to psychometric models. We extend the tests to linear mixed models here, detailing their implementation and performance when the data generating model is known. We also include an empirical example illustrating the tests' use in practice.

stat.ME↗