arXiv subjects

Xi Wu

Publications and source records attributed to Xi Wu.

At least 109 records · Page 6Linked to original sources

arXiv2020

Category-wise Attack: Transferable Adversarial Examples for Anchor Free Object Detection

Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbations can completely change the classification results. Their vulnerability has led to a surge of research in this direction. However, most works dedicated to attacking anchor-based object detection models. In this work, we aim to present an effective and efficient algorithm to generate adversarial examples to attack anchor-free object models based on two approaches. First, we conduct category-wise instead of instance-wise attacks on the object detectors. Second, we leverage the high-level semantic information to generate the adversarial examples. Surprisingly, the generated adversarial examples it not only able to effectively attack the targeted anchor-free object detector but also to be transferred to attack other object detectors, even anchor-based detectors such as Faster R-CNN.

cs.CV↗

arXiv2020

Representation Bayesian Risk Decompositions and Multi-Source Domain Adaptation

We consider representation learning (hypothesis class $\mathcal{H} = \mathcal{F}\circ\mathcal{G}$) where training and test distributions can be different. Recent studies provide hints and failure examples for domain invariant representation learning, a common approach for this problem, but the explanations provided are somewhat different and do not provide a unified picture. In this paper, we provide new decompositions of risk which give finer-grained explanations and clarify potential generalization issues. For Single-Source Domain Adaptation, we give an exact decomposition (an equality) of the target risk, via a natural hybrid argument, as sum of three factors: (1) source risk, (2) representation conditional label divergence, and (3) representation covariate shift. We derive a similar decomposition for the Multi-Source case. These decompositions reveal factors (2) and (3) as the precise reasons for failure to generalize. For example, we demonstrate that domain adversarial neural networks (DANN) attempt to regularize for (3) but miss (2), while a recent technique Invariant Risk Minimization (IRM) attempts to account for (2) but does not consider (3). We also verify our observations experimentally.

cs.LG↗

arXiv2020

TKNN formula for general lattice Hamiltonian in odd dimensions

Topological insulators in odd dimensions are characterized by topological numbers. We prove the well-known relation between the topological number given by the Chern character of the Berry curvature and the Chern-Simons level of the low energy effective action for a general class of Hamiltonians bilinear in the fermion with general U(1) gauge interactions including non-minimal couplings by an explicit calculation. A series of Ward-Takahashi identities are crucial to relate the Chern-Simons level to a winding number, which could then be directly reduced to Chern character of Berry curvature by carrying out the integral over the temporal momenta.

hep-lat↗

arXiv2020

TKNN formula for general Hamiltonian

hep-th↗

arXiv2020

Robust Multimodal Image Registration Using Deep Recurrent Reinforcement Learning

The crucial components of a conventional image registration method are the choice of the right feature representations and similarity measures. These two components, although elaborately designed, are somewhat handcrafted using human knowledge. To this end, these two components are tackled in an end-to-end manner via reinforcement learning in this work. Specifically, an artificial agent, which is composed of a combined policy and value network, is trained to adjust the moving image toward the right direction. We train this network using an asynchronous reinforcement learning algorithm, where a customized reward function is also leveraged to encourage robust image registration. This trained network is further incorporated with a lookahead inference to improve the registration capability. The advantage of this algorithm is fully demonstrated by our superior performance on clinical MR and CT image pairs to other state-of-the-art medical image registration methods.

cs.CV↗

arXiv2020

Quantum Hall conductivity in the presence of interactions

We discuss quantum Hall effect in the presence of arbitrary pair interactions between electrons. It is shown that irrespective of the interaction strength the Hall conductivity is given by the filling fraction of Landau levels averaged over the ground state of the system. This conclusion remains valid for both integer and fractional quantum Hall effect.

cond-mat.mes-hall↗

arXiv2019

Rearchitecting Classification Frameworks For Increased Robustness

While generalizing well over natural inputs, neural networks are vulnerable to adversarial inputs. Existing defenses against adversarial inputs have largely been detached from the real world. These defenses also come at a cost to accuracy. Fortunately, there are invariances of an object that are its salient features; when we break them it will necessarily change the perception of the object. We find that applying invariants to the classification task makes robustness and accuracy feasible together. Two questions follow: how to extract and model these invariances? and how to design a classification paradigm that leverages these invariances to improve the robustness accuracy trade-off? The remainder of the paper discusses solutions to the aformenetioned questions.

cs.LG↗

arXiv2019

Robust Attribution Regularization

An emerging problem in trustworthy machine learning is to train models that produce robust interpretations for their predictions. We take a step towards solving this problem through the lens of axiomatic attribution of neural networks. Our theory is grounded in the recent work, Integrated Gradients (IG), in axiomatically attributing a neural network's output change to its input change. We propose training objectives in classic robust optimization models to achieve robust IG attributions. Our objectives give principled generalizations of previous objectives designed for robust predictions, and they naturally degenerate to classic soft-margin training for one-layer neural networks. We also generalize previous theory and prove that the objectives for different robust optimization models are closely related. Experiments demonstrate the effectiveness of our method, and also point to intriguing problems which hint at the need for better optimization techniques or better neural network architectures for robust attribution training.

cs.LG↗

arXiv2019

Discovery of Bias and Strategic Behavior in Crowdsourced Performance Assessment

With the industry trend of shifting from a traditional hierarchical approach to flatter management structure, crowdsourced performance assessment gained mainstream popularity. One fundamental challenge of crowdsourced performance assessment is the risks that personal interest can introduce distortions of facts, especially when the system is used to determine merit pay or promotion. In this paper, we developed a method to identify bias and strategic behavior in crowdsourced performance assessment, using a rich dataset collected from a professional service firm in China. We find a pattern of "discriminatory generosity" on the part of peer evaluation, where raters downgrade their peer coworkers who have passed objective promotion requirements while overrating their peer coworkers who have not yet passed. This introduces two types of biases: the first aimed against more competent competitors, and the other favoring less eligible peers which can serve as a mask of the first bias. This paper also aims to bring angles of fairness-aware data mining to talent and management computing. Historical decision records, such as performance ratings, often contain subjective judgment which is prone to bias and strategic behavior. For practitioners of predictive talent analytics, it is important to investigate potential bias and strategic behavior underlying historical decision records.

cs.LG↗

arXiv2019

Towards Understanding Limitations of Pixel Discretization Against Adversarial Attacks

Wide adoption of artificial neural networks in various domains has led to an increasing interest in defending adversarial attacks against them. Preprocessing defense methods such as pixel discretization are particularly attractive in practice due to their simplicity, low computational overhead, and applicability to various systems. It is observed that such methods work well on simple datasets like MNIST, but break on more complicated ones like ImageNet under recently proposed strong white-box attacks. To understand the conditions for success and potentials for improvement, we study the pixel discretization defense method, including more sophisticated variants that take into account the properties of the dataset being discretized. Our results again show poor resistance against the strong attacks. We analyze our results in a theoretical framework and offer strong evidence that pixel discretization is unlikely to work on all but the simplest of the datasets. Furthermore, our arguments present insights why some other preprocessing defenses may be insecure.

cs.CR↗

arXiv2019

Tuple-oriented Compression for Large-scale Mini-batch Stochastic Gradient Descent

Data compression is a popular technique for improving the efficiency of data processing workloads such as SQL queries and more recently, machine learning (ML) with classical batch gradient methods. But the efficacy of such ideas for mini-batch stochastic gradient descent (MGD), arguably the workhorse algorithm of modern ML, is an open question. MGD's unique data access pattern renders prior art, including those designed for batch gradient methods, less effective. We fill this crucial research gap by proposing a new lossless compression scheme we call tuple-oriented compression (TOC) that is inspired by an unlikely source, the string/text compression scheme Lempel-Ziv-Welch, but tailored to MGD in a way that preserves tuple boundaries within mini-batches. We then present a suite of novel compressed matrix operation execution techniques tailored to the TOC compression scheme that operate directly over the compressed data representation and avoid decompression overheads. An extensive empirical evaluation with real-world datasets shows that TOC consistently achieves substantial compression ratios by up to 51x and reduces runtimes for MGD workloads by up to 10.2x in popular ML systems.

cs.LG↗

arXiv2018

Reinforcing Adversarial Robustness using Model Confidence Induced by Adversarial Training

In this paper we study leveraging confidence information induced by adversarial training to reinforce adversarial robustness of a given adversarially trained model. A natural measure of confidence is $\|F({\bf x})\|_\infty$ (i.e. how confident $F$ is about its prediction?). We start by analyzing an adversarial training formulation proposed by Madry et al.. We demonstrate that, under a variety of instantiations, an only somewhat good solution to their objective induces confidence to be a discriminator, which can distinguish between right and wrong model predictions in a neighborhood of a point sampled from the underlying distribution. Based on this, we propose Highly Confident Near Neighbor (${\tt HCNN}$), a framework that combines confidence information and nearest neighbor search, to reinforce adversarial robustness of a base model. We give algorithms in this framework and perform a detailed empirical study. We report encouraging experimental results that support our analysis, and also discuss problems we observed with existing adversarial training.

cs.LG↗

arXiv2018

Time-reversed magnetically controlled perturbation (TRMCP) optical focusing inside scattering media

Manipulating and focusing light deep inside biological tissue and tissue-like complex media has been desired for long yet considered challenging. One feasible strategy is through optical wavefront engineering, where the optical scattering-induced phase distortions are time reversed or pre-compensated so that photons travel along different optical paths interfere constructively at the targeted position within a scattering medium. To define the targeted position, an internal guidestar is needed to guide or provide a feedback for wavefront engineering. It could be injected or embedded probes such as fluorescence or nonlinear microspheres, ultrasonic modulation, as well as absorption perturbation. Here we propose to use a magnetically controlled optical absorbing microsphere as the internal guidestar. Using a digital optical phase conjugation system, we obtained sharp optical focusing within scattering media through time-reversing the scattered light perturbed by the magnetic microshpere. Since the object is magnetically controlled, dynamic optical focusing is allowed with a relatively large field-of-view by scanning the magnetic field externally. Moreover, the magnetic microsphere can be packaged with an organic membrane, using biological or chemical means to serve as a carrier. Therefore the technique may find particular applications for enhanced targeted drug delivery, and imaging and photoablation of angiogenic vessels in tumours.

physics.optics↗

arXiv2017

An improved design method for conventional straight dipole magnets

The standard design method for conventional straight dipole magnets is improved in this paper. The good field region is not symmetric with respect to the magnet mechanical center, and its width is not enlarged to include the beam sagitta. The integrated field quality is obtained by integrating the field along nominal beam paths. 2D and 3D design procedures of the improved design method are introduced, and two application examples of straight dipole magnets are presented. It is shown that the differences in integrated field quality between different field integration paths cannot be neglected. Compared with the traditional design method of straight dipole magnets, the advantage of the improved method is that the integrated field quality is accurate; the pole width, magnet dimension and weight of a straight dipole magnet can be reduced.

physics.acc-ph↗

arXiv2017

Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics

While significant progress has been made separately on analytics systems for scalable stochastic gradient descent (SGD) and private SGD, none of the major scalable analytics frameworks have incorporated differentially private SGD. There are two inter-related issues for this disconnect between research and practice: (1) low model accuracy due to added noise to guarantee privacy, and (2) high development and runtime overhead of the private algorithms. This paper takes a first step to remedy this disconnect and proposes a private SGD algorithm to address \emph{both} issues in an integrated manner. In contrast to the white-box approach adopted by previous work, we revisit and use the classical technique of {\em output perturbation} to devise a novel "bolt-on" approach to private SGD. While our approach trivially addresses (2), it makes (1) even more challenging. We address this challenge by providing a novel analysis of the $L_2$-sensitivity of SGD, which allows, under the same privacy guarantees, better convergence of SGD when only a constant number of passes can be made over the data. We integrate our algorithm, as well as other state-of-the-art differentially private SGD, into Bismarck, a popular scalable SGD-based analytics system on top of an RDBMS. Extensive experiments show that our algorithm can be easily integrated, incurs virtually no overhead, scales well, and most importantly, yields substantially better (up to 4X) test accuracy than the state-of-the-art algorithms on many real datasets.

cs.LG↗

arXiv2017

A Coordination Language for Databases

We present a coordination language for the modeling of distributed database applications. The language, baptized Klaim-DB, borrows the concepts of localities and nets of the coordination language Klaim but re-incarnates the tuple spaces of Klaim as databases. It provides high-level abstractions and primitives for the access and manipulation of structured data, with integrity and atomicity considerations. We present the formal semantics of Klaim-DB and develop a type system that avoids potential runtime errors such as certain evaluation errors and mismatches of data format in tables, which are monitored in the semantics. The use of the language is illustrated in a scenario where the sales from different branches of a chain of department stores are aggregated from their local databases. Raising the abstraction level and encapsulating integrity checks in the language primitives have benefited the modeling task considerably.

cs.LO↗

arXiv2017

Edge-of-edge States

We study an exotic state which is localized only at an intersection of edges of a topological material. This "edge-of-edge" state is shown to exist generically. We construct explicitly generic edge-of-edge states in 5-dimensional Weyl semimetals and their dimensional reductions, such as 4-dimensional topological insulators of class A and 3-dimensional chiral topological insulators of class AIII. The existence of the edge-of-edge state is due to a topological charge of the edge states. The notion of the Berry connection is generalized to include the space of all possible boundary conditions, where Chern-Simons forms are shown to be nontrivial.

cond-mat.mes-hall↗

arXiv2017

Boundary Conditions of Weyl Semimetals

We find that generic boundary conditions of Weyl semimetal is dictated by only a single real parameter, in the continuum limit. We determine how the energy dispersions (the Fermi arcs) and the wave functions of edge states depend on this parameter. Lattice models are found to be consistent with our generic observation. Furthermore, the enhanced parameter space of the boundary condition is shown to support a novel topological number.

cond-mat.mes-hall↗