arXiv subjects

Xiao Han

Publications and source records attributed to Xiao Han.

At least 163 records · Page 9Linked to original sources

arXiv2022

Large-Scale Product Retrieval with Weakly Supervised Representation Learning

Large-scale weakly supervised product retrieval is a practically useful yet computationally challenging problem. This paper introduces a novel solution for the eBay Visual Search Challenge (eProduct) held at the Ninth Workshop on Fine-Grained Visual Categorisation workshop (FGVC9) of CVPR 2022. This competition presents two challenges: (a) E-commerce is a drastically fine-grained domain including many products with subtle visual differences; (b) A lacking of target instance-level labels for model training, with only coarse category labels and product titles available. To overcome these obstacles, we formulate a strong solution by a set of dedicated designs: (a) Instead of using text training data directly, we mine thousands of pseudo-attributes from product titles and use them as the ground truths for multi-label classification. (b) We incorporate several strong backbones with advanced training recipes for more discriminative representation learning. (c) We further introduce a number of post-processing techniques including whitening, re-ranking and model ensemble for retrieval enhancement. By achieving 71.53% MAR, our solution "Involution King" achieves the second position on the leaderboard.

cs.CV↗

arXiv2022

FashionViL: Fashion-Focused Vision-and-Language Representation Learning

Large-scale Vision-and-Language (V+L) pre-training for representation learning has proven to be effective in boosting various downstream V+L tasks. However, when it comes to the fashion domain, existing V+L methods are inadequate as they overlook the unique characteristics of both the fashion V+L data and downstream tasks. In this work, we propose a novel fashion-focused V+L representation learning framework, dubbed as FashionViL. It contains two novel fashion-specific pre-training tasks designed particularly to exploit two intrinsic attributes with fashion V+L data. First, in contrast to other domains where a V+L data point contains only a single image-text pair, there could be multiple images in the fashion domain. We thus propose a Multi-View Contrastive Learning task for pulling closer the visual representation of one image to the compositional multimodal representation of another image+text. Second, fashion text (e.g., product description) often contains rich fine-grained concepts (attributes/noun phrases). To exploit this, a Pseudo-Attributes Classification task is introduced to encourage the learned unimodal (visual/textual) representations of the same concept to be adjacent. Further, fashion V+L tasks uniquely include ones that do not conform to the common one-stream or two-stream architectures (e.g., text-guided image retrieval). We thus propose a flexible, versatile V+L model architecture consisting of a modality-agnostic Transformer so that it can be flexibly adapted to any downstream tasks. Extensive experiments show that our FashionViL achieves a new state of the art across five downstream tasks. Code is available at https://github.com/BrandonHanx/mmf.

cs.CV↗

arXiv2022

Large-Scale Privacy-Preserving Network Embedding against Private Link Inference Attacks

Network embedding represents network nodes by a low-dimensional informative vector. While it is generally effective for various downstream tasks, it may leak some private information of networks, such as hidden private links. In this work, we address a novel problem of privacy-preserving network embedding against private link inference attacks. Basically, we propose to perturb the original network by adding or removing links, and expect the embedding generated on the perturbed network can leak little information about private links but hold high utility for various downstream tasks. Towards this goal, we first propose general measurements to quantify privacy gain and utility loss incurred by candidate network perturbations; we then design a PPNE framework to identify the optimal perturbation solution with the best privacy-utility trade-off in an iterative way. Furthermore, we propose many techniques to accelerate PPNE and ensure its scalability. For instance, as the skip-gram embedding methods including DeepWalk and LINE can be seen as matrix factorization with closed form embedding results, we devise efficient privacy gain and utility loss approximation methods to avoid the repetitive time-consuming embedding training for every candidate network perturbation in each iteration. Experiments on real-life network datasets (with up to millions of nodes) verify that PPNE outperforms baselines by sacrificing less utility and obtaining higher privacy protection.

cs.LG↗

arXiv2022

UIGR: Unified Interactive Garment Retrieval

Interactive garment retrieval (IGR) aims to retrieve a target garment image based on a reference garment image along with user feedback on what to change on the reference garment. Two IGR tasks have been studied extensively: text-guided garment retrieval (TGR) and visually compatible garment retrieval (VCR). The user feedback for the former indicates what semantic attributes to change with the garment category preserved, while the category is the only thing to be changed explicitly for the latter, with an implicit requirement on style preservation. Despite the similarity between these two tasks and the practical need for an efficient system tackling both, they have never been unified and modeled jointly. In this paper, we propose a Unified Interactive Garment Retrieval (UIGR) framework to unify TGR and VCR. To this end, we first contribute a large-scale benchmark suited for both problems. We further propose a strong baseline architecture to integrate TGR and VCR in one model. Extensive experiments suggest that unifying two tasks in one framework is not only more efficient by requiring a single model only, it also leads to better performance. Code and datasets are available at https://github.com/BrandonHanx/CompFashion.

cs.CV↗

arXiv2022

Hopf algebroids from noncommutative bundles

We present two classes of examples of Hopf algebroids associated with noncommutative principal bundles. The first comes from deforming the principal bundle while leaving unchanged the structure Hopf algebra. The second is related to deforming a quantum homogeneous space; this needs a careful deformation of the structure Hopf algebra in order to preserve the compatibilities between the Hopf algebra operations.

math.QA↗

arXiv2021

HyObscure: Hybrid Obscuring for Privacy-Preserving Data Publishing

Minimizing privacy leakage while ensuring data utility is a critical problem to data holders in a privacy-preserving data publishing task. Most prior research concerns only with one type of data and resorts to a single obscuring method, \eg, obfuscation or generalization, to achieve a privacy-utility tradeoff, which is inadequate for protecting real-life heterogeneous data and is hard to defend ever-growing machine learning based inference attacks. This work takes a pilot study on privacy-preserving data publishing when both generalization and obfuscation operations are employed for heterogeneous data protection. To this end, we first propose novel measures for privacy and utility quantification and formulate the hybrid privacy-preserving data obscuring problem to account for the joint effect of generalization and obfuscation. We then design a novel hybrid protection mechanism called HyObscure, to cross-iteratively optimize the generalization and obfuscation operations for maximum privacy protection under a certain utility guarantee. The convergence of the iterative process and the privacy leakage bound of HyObscure are also provided in theory. Extensive experiments demonstrate that HyObscure significantly outperforms a variety of state-of-the-art baseline methods when facing various inference attacks under different scenarios. HyObscure also scales linearly to the data size and behaves robustly with varying key parameters.

cs.CR↗

arXiv2021

CoDiM: Learning with Noisy Labels via Contrastive Semi-Supervised Learning

Labels are costly and sometimes unreliable. Noisy label learning, semi-supervised learning, and contrastive learning are three different strategies for designing learning processes requiring less annotation cost. Semi-supervised learning and contrastive learning have been recently demonstrated to improve learning strategies that address datasets with noisy labels. Still, the inner connections between these fields as well as the potential to combine their strengths together have only started to emerge. In this paper, we explore further ways and advantages to fuse them. Specifically, we propose CSSL, a unified Contrastive Semi-Supervised Learning algorithm, and CoDiM (Contrastive DivideMix), a novel algorithm for learning with noisy labels. CSSL leverages the power of classical semi-supervised learning and contrastive learning technologies and is further adapted to CoDiM, which learns robustly from multiple types and levels of label noise. We show that CoDiM brings consistent improvements and achieves state-of-the-art results on multiple benchmarks.

cs.LG↗

arXiv2021

Microgroove convexity is critical for robust gaseous layers on hierarchically-structured superhydrophobic surfaces

Gaseous layers (plastrons) trapped on the surfaces of immersed hydrophobic surfaces are critical for their function. Fibrillar morphologies offer a natural pathway, yet they are limited to a narrow range of liquid-surface systems and are vulnerable to pressure fluctuations that irreversibly destroy the plastron. Inspired by the convexly grooved surfaces of water fern (Salvinia) leaves that support their fibrous outgrowths, we study the plastron formation on 3D-printed dual-scale surfaces with elliptical interconnected microgrooves. The groove curvature stabilizes a seed gas layer (SGL) that facilitates plastron formation and restoration for all immersed hydrophobic surfaces. Computations and theoretical calculations reveal that the SGL storage capacity that sets the plastron robustness follows from the liquid menisci adaption to the groove geometry and pressure, and it can be further tuned using separated grooves. Our study highlights groove convexity as a key morphological feature for the design of multi-scale immersed surfaces for robust superhydrophobicity.

cond-mat.soft↗

arXiv2021

Gauge groups and bialgebroids

We study the Ehresmann--Schauenburg bialgebroid of a noncommutative principal bundle as a quantization of the gauge groupoid of a classical principal bundle. We show that the gauge group of the noncommutative bundle is isomorphic to the group of bisections of the bialgebroid, and we give a crossed module structure for the bisections and the automorphisms of the bialgebroid. Examples illustrating these constructions include: Galois objects of Taft algebras, a monopole bundle over a quantum spheres and a not faithfully flat Hopf--Galois extension of commutative algebras. The latter two examples have in fact a structure of Hopf algebroid for a suitable invertible antipode.

math.QA↗

arXiv2021

Text-Based Person Search with Limited Data

Text-based person search (TBPS) aims at retrieving a target person from an image gallery with a descriptive text query. Solving such a fine-grained cross-modal retrieval task is challenging, which is further hampered by the lack of large-scale datasets. In this paper, we present a framework with two novel components to handle the problems brought by limited data. Firstly, to fully utilize the existing small-scale benchmarking datasets for more discriminative feature learning, we introduce a cross-modal momentum contrastive learning framework to enrich the training data for a given mini-batch. Secondly, we propose to transfer knowledge learned from existing coarse-grained large-scale datasets containing image-text pairs from drastically different problem domains to compensate for the lack of TBPS training data. A transfer learning method is designed so that useful information can be transferred despite the large domain gap. Armed with these components, our method achieves new state of the art on the CUHK-PEDES dataset with significant improvements over the prior art in terms of Rank-1 and mAP. Our code is available at https://github.com/BrandonHanx/TextReID.

cs.CV↗

arXiv2021

SIMPLE: Statistical Inference on Membership Profiles in Large Networks

Network data is prevalent in many contemporary big data applications in which a common interest is to unveil important latent links between different pairs of nodes. Yet a simple fundamental question of how to precisely quantify the statistical uncertainty associated with the identification of latent links still remains largely unexplored. In this paper, we propose the method of statistical inference on membership profiles in large networks (SIMPLE) in the setting of degree-corrected mixed membership model, where the null hypothesis assumes that the pair of nodes share the same profile of community memberships. In the simpler case of no degree heterogeneity, the model reduces to the mixed membership model for which an alternative more robust test is also proposed. Both tests are of the Hotelling-type statistics based on the rows of empirical eigenvectors or their ratios, whose asymptotic covariance matrices are very challenging to derive and estimate. Nevertheless, their analytical expressions are unveiled and the unknown covariance matrices are consistently estimated. Under some mild regularity conditions, we establish the exact limiting distributions of the two forms of SIMPLE test statistics under the null hypothesis and contiguous alternative hypothesis. They are the chi-square distributions and the noncentral chi-square distributions, respectively, with degrees of freedom depending on whether the degrees are corrected or not. We also address the important issue of estimating the unknown number of communities and establish the asymptotic properties of the associated test statistics. The advantages and practical utility of our new procedures in terms of both size and power are demonstrated through several simulation examples and real network applications.

stat.ME↗

arXiv2021

On the gauge group of Galois objects

We study the Ehresmann--Schauenburg bialgebroid of a noncommutative principal bundle as a quantization of the classical gauge groupoid of a principal bundle. When the base algebra is in the centre of the total space algebra, the gauge group of the noncommutative principal bundle is isomorphic to the group of bisections of the bialgebroid. In particular we consider Galois objects (non-trivial noncommutative bundles over a point in a sense) for which the bialgebroid is a Hopf algebra. For these we give a crossed module structure for the bisections and the automorphisms of the bialgebroid. Examples include Galois objects of group Hopf algebras and of Taft algebras.

math.QA↗

arXiv2021

Misbehavior Detection in Wi-Fi/LTE Coexistence over Unlicensed Bands

We consider the problem of fair coexistence between LTE and Wi-Fi systems in the unlicensed 5 GHz U-NII bands. We focus on the misbehavior opportunities due to the heterogeneity in channel access mechanism and the lack of a common control plane. We define selfish misbehavior strategies for the LTE that yield an unfair share of the spectrum resources. Such strategies are based on manipulating the operational parameters of the LTE-LAA standard, namely the backoff mechanism, the traffic class parameters, the clear channel access (CCA) threshold, and others. Prior methods for detecting misbehavior in homogeneous settings are not applicable in a spectrum sharing scenario because the devices of one system cannot decode the transmissions of another. We develop implicit sensing techniques that can accurately estimate the operational parameters of LTE transmissions under various topological scenarios and {\em without decoding.} These techniques apply correlation-based signal detection to infer the required information. Our techniques are validated through experiments on a USRP testbed. We further apply a statistical inference framework for determining deviations of the LTE behavior from the coexistence etiquette. By characterizing the detection and false alarm probabilities, we show that our framework yields high detection accuracy at a very low false alarm rate. Although our methods focus on detecting misbehavior of the LTE system, they can be generalized to other coexistence scenarios.

cs.NI↗

arXiv2020

Label Confusion Learning to Enhance Text Classification Models

Representing a true label as a one-hot vector is a common practice in training text classification models. However, the one-hot representation may not adequately reflect the relation between the instances and labels, as labels are often not completely independent and instances may relate to multiple labels in practice. The inadequate one-hot representations tend to train the model to be over-confident, which may result in arbitrary prediction and model overfitting, especially for confused datasets (datasets with very similar labels) or noisy datasets (datasets with labeling errors). While training models with label smoothing (LS) can ease this problem in some degree, it still fails to capture the realistic relation among labels. In this paper, we propose a novel Label Confusion Model (LCM) as an enhancement component to current popular text classification models. LCM can learn label confusion to capture semantic overlap among labels by calculating the similarity between instances and labels during training and generate a better label distribution to replace the original one-hot label vector, thus improving the final classification performance. Extensive experiments on five text classification benchmark datasets reveal the effectiveness of LCM for several widely used deep learning classification models. Further experiments also verify that LCM is especially helpful for confused or noisy datasets and superior to the label smoothing method.

cs.CL↗

arXiv2020

Federated Crowdsensing: Framework and Challenges

Crowdsensing is a promising sensing paradigm for smart city applications (e.g., traffic and environment monitoring) with the prevalence of smart mobile devices and advanced network infrastructure. Meanwhile, as tasks are performed by individuals, privacy protection is one of the key issues in crowdsensing systems. Traditionally, to alleviate users' privacy concerns, noises are added to participants' sensitive data (e.g., participants' locations) through techniques such as differential privacy. However, this inevitably results in quality loss to the crowdsensing task. Recently, federated learning paradigm has been proposed, which aims to achieve privacy preservation in machine learning while ensuring that the learning quality suffers little or no loss. Inspired by the federated learning paradigm, this article studies how federated learning may benefit crowdsensing applications. In particular, we first propose a federated crowdsensing framework, which analyzes the privacy concerns of each crowdsensing stage (i.e., task creation, task assignment, task execution, and data aggregation) and discuss how federated learning techniques may take effect. Finally, we summarize key challenges and opportunities in federated crowdsensing.

cs.CR↗

arXiv2020

Asymptotic Theory of Eigenvectors for Random Matrices with Diverging Spikes

Characterizing the asymptotic distributions of eigenvectors for large random matrices poses important challenges yet can provide useful insights into a range of statistical applications. To this end, in this paper we introduce a general framework of asymptotic theory of eigenvectors (ATE) for large spiked random matrices with diverging spikes and heterogeneous variances, and establish the asymptotic properties of the spiked eigenvectors and eigenvalues for the scenario of the generalized Wigner matrix noise. Under some mild regularity conditions, we provide the asymptotic expansions for the spiked eigenvalues and show that they are asymptotically normal after some normalization. For the spiked eigenvectors, we establish asymptotic expansions for the general linear combination and further show that it is asymptotically normal after some normalization, where the weight vector can be arbitrary. We also provide a more general asymptotic theory for the spiked eigenvectors using the bilinear form. Simulation studies verify the validity of our new theoretical results. Our family of models encompasses many popularly used ones such as the stochastic block models with or without overlapping communities for network analysis and the topic models for text analysis, and our general theory can be exploited for statistical inference in these large-scale applications.

math.ST↗

arXiv2020

Microscope Based HER2 Scoring System

The overexpression of human epidermal growth factor receptor 2 (HER2) has been established as a therapeutic target in multiple types of cancers, such as breast and gastric cancers. Immunohistochemistry (IHC) is employed as a basic HER2 test to identify the HER2-positive, borderline, and HER2-negative patients. However, the reliability and accuracy of HER2 scoring are affected by many factors, such as pathologists' experience. Recently, artificial intelligence (AI) has been used in various disease diagnosis to improve diagnostic accuracy and reliability, but the interpretation of diagnosis results is still an open problem. In this paper, we propose a real-time HER2 scoring system, which follows the HER2 scoring guidelines to complete the diagnosis, and thus each step is explainable. Unlike the previous scoring systems based on whole-slide imaging, our HER2 scoring system is integrated into an augmented reality (AR) microscope that can feedback AI results to the pathologists while reading the slide. The pathologists can help select informative fields of view (FOVs), avoiding the confounding regions, such as DCIS. Importantly, we illustrate the intermediate results with membrane staining condition and cell classification results, making it possible to evaluate the reliability of the diagnostic results. Also, we support the interactive modification of selecting regions-of-interest, making our system more flexible in clinical practice. The collaboration of AI and pathologists can significantly improve the robustness of our system. We evaluate our system with 285 breast IHC HER2 slides, and the classification accuracy of 95\% shows the effectiveness of our HER2 scoring system.

eess.IV↗

arXiv2020

On Chemical Distance and Local Uniqueness of a Sufficiently Supercritical Finitary Random Interlacement

In this paper, we study geometric properties of the unique infinite cluster $Γ$ in a sufficiently supercritical Finitary Random Interlacements $\mathcal{FI}^{u,T}$ in $\mathbb{Z}^d, \ d\ge 3$. We prove that the chemical distance in $Γ$ is, with stretched exponentially high probability, of the same order as the Euclidean distance in $\mathbb{Z}^d$. This also implies a shape theorem parallel to those for Bernoulli percolation and random interlacements. We also prove local uniqueness of $\mathcal{FI}^{u,T}$, which says any two large clusters in $\mathcal{FI}^{u,T}$ "close to each other" will with stretched exponentially high probability be connected to each other within the same order of the distance between them.

math.PR↗