arXiv subjects

Wenchao Xu

Publications and source records attributed to Wenchao Xu.

At least 73 records · Page 4Linked to original sources

arXiv2023

MDENet: Multi-modal Dual-embedding Networks for Malware Open-set Recognition

Malware open-set recognition (MOSR) aims at jointly classifying malware samples from known families and detect the ones from novel unknown families, respectively. Existing works mostly rely on a well-trained classifier considering the predicted probabilities of each known family with a threshold-based detection to achieve the MOSR. However, our observation reveals that the feature distributions of malware samples are extremely similar to each other even between known and unknown families. Thus the obtained classifier may produce overly high probabilities of testing unknown samples toward known families and degrade the model performance. In this paper, we propose the Multi-modal Dual-Embedding Networks, dubbed MDENet, to take advantage of comprehensive malware features (i.e., malware images and malware sentences) from different modalities to enhance the diversity of malware feature space, which is more representative and discriminative for down-stream recognition. Last, to further guarantee the open-set recognition, we dually embed the fused multi-modal representation into one primary space and an associated sub-space, i.e., discriminative and exclusive spaces, with contrastive sampling and rho-bounded enclosing sphere regularizations, which resort to classification and detection, respectively. Moreover, we also enrich our previously proposed large-scaled malware dataset MAL-100 with multi-modal characteristics and contribute an improved version dubbed MAL-100+. Experimental results on the widely used malware dataset Mailing and the proposed MAL-100+ demonstrate the effectiveness of our method.

cs.CR↗

arXiv2023

Towards Unbiased Training in Federated Open-world Semi-supervised Learning

Federated Semi-supervised Learning (FedSSL) has emerged as a new paradigm for allowing distributed clients to collaboratively train a machine learning model over scarce labeled data and abundant unlabeled data. However, existing works for FedSSL rely on a closed-world assumption that all local training data and global testing data are from seen classes observed in the labeled dataset. It is crucial to go one step further: adapting FL models to an open-world setting, where unseen classes exist in the unlabeled data. In this paper, we propose a novel Federatedopen-world Semi-Supervised Learning (FedoSSL) framework, which can solve the key challenge in distributed and open-world settings, i.e., the biased training process for heterogeneously distributed unseen classes. Specifically, since the advent of a certain unseen class depends on a client basis, the locally unseen classes (exist in multiple clients) are likely to receive differentiated superior aggregation effects than the globally unseen classes (exist only in one client). We adopt an uncertainty-aware suppressed loss to alleviate the biased training between locally unseen and globally unseen classes. Besides, we enable a calibration module supplementary to the global aggregation to avoid potential conflicting knowledge transfer caused by inconsistent data distribution among different clients. The proposed FedoSSL can be easily adapted to state-of-the-art FL methods, which is also validated via extensive experiments on benchmarks and real-world datasets (CIFAR-10, CIFAR-100 and CINIC-10).

cs.LG↗

arXiv2023

DualMix: Unleashing the Potential of Data Augmentation for Online Class-Incremental Learning

Online Class-Incremental (OCI) learning has sparked new approaches to expand the previously trained model knowledge from sequentially arriving data streams with new classes. Unfortunately, OCI learning can suffer from catastrophic forgetting (CF) as the decision boundaries for old classes can become inaccurate when perturbated by new ones. Existing literature have applied the data augmentation (DA) to alleviate the model forgetting, while the role of DA in OCI has not been well understood so far. In this paper, we theoretically show that augmented samples with lower correlation to the original data are more effective in preventing forgetting. However, aggressive augmentation may also reduce the consistency between data and corresponding labels, which motivates us to exploit proper DA to boost the OCI performance and prevent the CF problem. We propose the Enhanced Mixup (EnMix) method that mixes the augmented samples and their labels simultaneously, which is shown to enhance the sample diversity while maintaining strong consistency with corresponding labels. Further, to solve the class imbalance problem, we design an Adaptive Mixup (AdpMix) method to calibrate the decision boundaries by mixing samples from both old and new classes and dynamically adjusting the label mixing ratio. Our approach is demonstrated to be effective on several benchmark datasets through extensive experiments, and it is shown to be compatible with other replay-based techniques.

cs.LG↗

arXiv2022

AirCon: Over-the-Air Consensus for Wireless Blockchain Networks

Blockchain has been deemed as a promising solution for providing security and privacy protection in the next-generation wireless networks. Large-scale concurrent access for massive wireless devices to accomplish the consensus procedure may consume prohibitive communication and computing resources, and thus may limit the application of blockchain in wireless conditions. As most existing consensus protocols are designed for wired networks, directly apply them for wireless users may exhaust their scarce spectrum and computing resources. In this paper, we propose AirCon, a byzantine fault-tolerant (BFT) consensus protocol for wireless users via the over-the-air computation. The novelty of AirCon is to take advantage of the intrinsic characteristic of the wireless channel and automatically achieve the consensus in the physical layer while receiving from the end users, which greatly reduces the communication and computational cost that would be caused by traditional consensus protocols. We implement the AirCon protocol integrated into an LTE system and provide solutions to the critical issues for over-the-air consensus implementation. Experimental results are provided to show the feasibility of the proposed protocol, and simulation results to show the performance of the AirCon protocol under different wireless conditions.

eess.SP↗

arXiv2022

FedTune: A Deep Dive into Efficient Federated Fine-Tuning with Pre-trained Transformers

Federated Learning (FL) is an emerging paradigm that enables distributed users to collaboratively and iteratively train machine learning models without sharing their private data. Motivated by the effectiveness and robustness of self-attention-based architectures, researchers are turning to using pre-trained Transformers (i.e., foundation models) instead of traditional convolutional neural networks in FL to leverage their excellent transfer learning capabilities. Despite recent progress, how pre-trained Transformer models play a role in FL remains obscure, that is, how to efficiently fine-tune these pre-trained models in FL and how FL users could benefit from this new paradigm. In this paper, we explore this issue and demonstrate that the fine-tuned Transformers achieve extraordinary performance on FL, and that the lightweight fine-tuning method facilitates a fast convergence rate and low communication costs. Concretely, we conduct a rigorous empirical study of three tuning methods (i.e., modifying the input, adding extra modules, and adjusting the backbone) using two types of pre-trained models (i.e., vision-language models and vision models) for FL. Our experiments show that 1) Fine-tuning the bias term of the backbone performs best when relying on a strong pre-trained model; 2) The vision-language model (e.g., CLIP) outperforms the pure vision model (e.g., ViT) and is more robust to the few-shot settings; 3) Compared to pure local training, FL with pre-trained models has a higher accuracy because it alleviates the problem of over-fitting. We will release our code and encourage further exploration of pre-trained Transformers and FL.

cs.LG↗

arXiv2022

PMR: Prototypical Modal Rebalance for Multimodal Learning

Multimodal learning (MML) aims to jointly exploit the common priors of different modalities to compensate for their inherent limitations. However, existing MML methods often optimize a uniform objective for different modalities, leading to the notorious "modality imbalance" problem and counterproductive MML performance. To address the problem, some existing methods modulate the learning pace based on the fused modality, which is dominated by the better modality and eventually results in a limited improvement on the worse modal. To better exploit the features of multimodal, we propose Prototypical Modality Rebalance (PMR) to perform stimulation on the particular slow-learning modality without interference from other modalities. Specifically, we introduce the prototypes that represent general features for each class, to build the non-parametric classifiers for uni-modal performance evaluation. Then, we try to accelerate the slow-learning modality by enhancing its clustering toward prototypes. Furthermore, to alleviate the suppression from the dominant modality, we introduce a prototype-based entropy regularization term during the early training stage to prevent premature convergence. Besides, our method only relies on the representations of each modality and without restrictions from model structures and fusion methods, making it with great application potential for various scenarios.

cs.LG↗

arXiv2022

Demystify Self-Attention in Vision Transformers from a Semantic Perspective: Analysis and Application

Self-attention mechanisms, especially multi-head self-attention (MSA), have achieved great success in many fields such as computer vision and natural language processing. However, many existing vision transformer (ViT) works simply inherent transformer designs from NLP to adapt vision tasks, while ignoring the fundamental difference between ``how MSA works in image and language settings''. Language naturally contains highly semantic structures that are directly interpretable by humans. Its basic unit (word) is discrete without redundant information, which readily supports interpretable studies on MSA mechanisms of language transformer. In contrast, visual data exhibits a fundamentally different structure: Its basic unit (pixel) is a natural low-level representation with significant redundancies in the neighbourhood, which poses obvious challenges to the interpretability of MSA mechanism in ViT. In this paper, we introduce a typical image processing technique, i.e., scale-invariant feature transforms (SIFTs), which maps low-level representations into mid-level spaces, and annotates extensive discrete keypoints with semantically rich information. Next, we construct a weighted patch interrelation analysis based on SIFT keypoints to capture the attention patterns hidden in patches with different semantic concentrations Interestingly, we find this quantitative analysis is not only an effective complement to the interpretability of MSA mechanisms in ViT, but can also be applied to 1) spurious correlation discovery and ``prompting'' during model inference, 2) and guided model pre-training acceleration. Experimental results on both applications show significant advantages over baselines, demonstrating the efficacy of our method.

cs.CV↗

arXiv2022

SigT: An Efficient End-to-End MIMO-OFDM Receiver Framework Based on Transformer

Multiple-input multiple-output and orthogonal frequency-division multiplexing (MIMO-OFDM) are the key technologies in 4G and subsequent wireless communication systems. Conventionally, the MIMO-OFDM receiver is performed by multiple cascaded blocks with different functions and the algorithm in each block is designed based on ideal assumptions of wireless channel distributions. However, these assumptions may fail in practical complex wireless environments. The deep learning (DL) method has the ability to capture key features from complex and huge data. In this paper, a novel end-to-end MIMO-OFDM receiver framework based on \textit{transformer}, named SigT, is proposed. By regarding the signal received from each antenna as a token of the transformer, the spatial correlation of different antennas can be learned and the critical zero-shot problem can be mitigated. Furthermore, the proposed SigT framework can work well without the inserted pilots, which improves the useful data transmission efficiency. Experiment results show that SigT achieves much higher performance in terms of signal recovery accuracy than benchmark methods, even in a low SNR environment or with a small number of training samples. Code is available at https://github.com/SigTransformer/SigT.

cs.NI↗

arXiv2022

PromptFL: Let Federated Participants Cooperatively Learn Prompts Instead of Models -- Federated Learning in Age of Foundation Model

Quick global aggregation of effective distributed parameters is crucial to federated learning (FL), which requires adequate bandwidth for parameters communication and sufficient user data for local training. Otherwise, FL may cost excessive training time for convergence and produce inaccurate models. In this paper, we propose a brand-new FL framework, PromptFL, that replaces the federated model training with the federated prompt training, i.e., let federated participants train prompts instead of a shared model, to simultaneously achieve the efficient global aggregation and local training on insufficient data by exploiting the power of foundation models (FM) in a distributed way. PromptFL ships an off-the-shelf FM, i.e., CLIP, to distributed clients who would cooperatively train shared soft prompts based on very few local data. Since PromptFL only needs to update the prompts instead of the whole model, both the local training and the global aggregation can be significantly accelerated. And FM trained over large scale data can provide strong adaptation capability to distributed users tasks with the trained soft prompts. We empirically analyze the PromptFL via extensive experiments, and show its superiority in terms of system feasibility, user privacy, and performance.

cs.LG↗

arXiv2022

Experimental study of humidity influence on triboelectric charging of particle-laden duct flows

Electrostatic charge on powders arises during pneumatic transport due to particle-particle and particle-surface interactions via triboelectrification. This is a potential threat to the safety of industrial productions and the source of numerous fires and dust explosions in the past. Triboelectric charges are affected by environmental conditions, such as ambient temperature and relative humidity. In this work, we experimentally investigated the influence of ambient humidity on the particle charge of gas-solid flows in a square-shaped duct. Monodisperse PMMA particles are fed into a fully developed airflow in a PMMA duct and then pass through a metallic duct section. The charge of particles is measured at the outlet of the metallic duct via a Faraday cup. By measuring the electrostatic charge under various environmental conditions, we observed that the electrostatic charge first increases with the humidity and then decreases when the humidity becomes higher.

physics.flu-dyn↗

arXiv2022

Layer-wised Model Aggregation for Personalized Federated Learning

Personalized Federated Learning (pFL) not only can capture the common priors from broad range of distributed data, but also support customized models for heterogeneous clients. Researches over the past few years have applied the weighted aggregation manner to produce personalized models, where the weights are determined by calibrating the distance of the entire model parameters or loss values, and have yet to consider the layer-level impacts to the aggregation process, leading to lagged model convergence and inadequate personalization over non-IID datasets. In this paper, we propose a novel pFL training framework dubbed Layer-wised Personalized Federated learning (pFedLA) that can discern the importance of each layer from different clients, and thus is able to optimize the personalized model aggregation for clients with heterogeneous data. Specifically, we employ a dedicated hypernetwork per client on the server side, which is trained to identify the mutual contribution factors at layer granularity. Meanwhile, a parameterized mechanism is introduced to update the layer-wised aggregation weights to progressively exploit the inter-user similarity and realize accurate model personalization. Extensive experiments are conducted over different models and learning tasks, and we show that the proposed methods achieve significantly higher performance than state-of-the-art pFL methods.

cs.DC↗

arXiv2022

Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representations

Recently, the enactment of privacy regulations has promoted the rise of the machine unlearning paradigm. Existing studies of machine unlearning mainly focus on sample-wise unlearning, such that a learnt model will not expose user's privacy at the sample level. Yet we argue that such ability of selective removal should also be presented at the attribute level, especially for the attributes irrelevant to the main task, e.g., whether a person recognized in a face recognition system wears glasses or the age range of that person. Through a comprehensive literature review, it is found that existing studies on attribute-related problems like fairness and de-biasing learning cannot address the above concerns properly. To bridge this gap, we propose a paradigm of selectively removing input attributes from feature representations which we name `attribute unlearning'. In this paradigm, certain attributes will be accurately captured and detached from the learned feature representations at the stage of training, according to their mutual information. The particular attributes will be progressively eliminated along with the training procedure towards convergence, while the rest of attributes related to the main task are preserved for achieving competitive model performance. Considering the computational complexity during the training process, we not only give a theoretically approximate training method, but also propose an acceleration scheme to speed up the training process. We validate our method by spanning several datasets and models and demonstrate that our design can preserve model fidelity and reach prevailing unlearning efficacy with high efficiency. The proposed unlearning paradigm builds a foundation for future machine unlearning system and will become an essential component of the latest privacy-related legislation.

cs.LG↗

arXiv2022

From Model Selection to Model Averaging: A Comparison for Nested Linear Models

Model selection (MS) and model averaging (MA) are two popular approaches when having many candidate models. Theoretically, the estimation risk of an oracle MA is not larger than that of an oracle MS because the former one is more flexible, but a foundational issue is: does MA offer a {\it substantial} improvement over MS? Recently, a seminal work: Peng and Yang (2021), has answered this question under nested models with linear orthonormal series expansion. In the current paper, we further reply this question under linear nested regression models. Especially, a more general nested framework, heteroscedastic and autocorrelated random errors, and sparse coefficients are allowed in the current paper, which is more common in practice. In addition, we further compare MAs with different weight sets. Simulation studies support the theoretical findings in a variety of settings.

math.ST↗

arXiv2022

Simulation of PMMA powder flow electrification using a new charging model based on single-particle experiments

Thus far, simulations have failed to predict accurately electrostatic powder charging during pneumatic transport. We advanced the modeling of powder flow charging by a three-part study: first, we shot individual particles on a metal target and measured the exchanged charge. Second, based on these results, we formulated an empirical model and implemented it in our CFD tool. Using this tool, we performed large-eddy simulations of the powder flow through a square duct with a cylindrical obstacle inside. Finally, we compared the simulations to measurements in our pneumatic conveying test rig. The simulations successfully predicted the charging of powder consisting of monodisperse particles of a size of 200 $μ$m. Contrary to the usual procedure for this type of simulation, the tool requires no tuning of any parameters. According to our simulations, the powder mostly charged when hitting the cylindrical obstacle. The contacts led to bipolar charge distributions.

physics.flu-dyn↗

arXiv2021

Fast Preparation and Detection of a Rydberg Qubit using Atomic Ensembles

We demonstrate a new approach for fast preparation, manipulation, and collective readout of an atomic Rydberg-state qubit. By making use of Rydberg blockade inside a small atomic ensemble, we prepare a single qubit within 3~$μ$s with a success probability of $F_p=0.93 \pm 0.02$, rotate it, and read out its state in $6$ $μs$ with a single-shot fidelity of $F_d=0.92 \pm 0.04$. The ensemble-assisted detection is $10^3$ times faster than imaging of a single atom with the same optical resolution, and enables fast repeated non-destructive measurement. We observe qubit coherence times of 15~$μ$s, much longer than the $π$ rotation time of 90~ns. Potential applications ranging from faster quantum information processing in atom arrays to efficient implementation of quantum error correction are discussed.

quant-ph↗

arXiv2019

Repulsive photons in a quantum nonlinear medium

The ability to control strongly interacting light quanta (photons) is of central importance in quantum science and engineering. Recently it was shown that such strong interactions can be engineered in specially prepared quantum optical systems. Here, we demonstrate a method for coherent control of strongly interacting photons, extending quantum nonlinear optics into the domain of repulsive photons. This is achieved by coherently coupling photons to several atomic states, including strongly interacting Rydberg levels in a cold Rubidium gas. Using this approach we demonstrate both repulsive and attractive interactions between individual photons and characterize them by the measured two- and three-photon correlation functions. For the repulsive case, we demonstrate signatures of interference and self ordering from three-photon measurements. These observations open a route to study strongly interacting dissipative systems and quantum matter composed of light such as a crystal of individual photons.

physics.atom-ph↗

arXiv2018

Bad-Metal Relaxation Dynamics in a Fermi Lattice Gas

We report the discovery of phenomena consistent with bad-metal relaxation dynamics in the metallic regime of an optical-lattice Hubbard model. The transport lifetime induced by inter-particle scattering for a mass current of atoms excited by stimulated Raman transitions is measured, and the corresponding analog of resistivity is inferred. By exploring a range of temperature, we demonstrate incompatibility with weak-scattering theory and a key characteristic of bad metals: anomalous resistivity scaling consistent with $T$-linear behavior. We also observe the onset of two behaviors---incoherent transport and the approach to the Mott-Ioffe-Regel limit---associated with bad metals. The interaction and temperature scaling of resistivity are verified to be consistent with dynamic mean-field theory (DMFT) predictions of a bad metal, which is associated with the reduction of quasiparticle weight by strong interactions.

cond-mat.quant-gas↗

arXiv2018

Air-Ground Integrated Mobile Edge Networks: Architecture, Challenges and Opportunities

The ever-increasing mobile data demands have posed significant challenges in the current radio access networks, while the emerging computation-heavy Internet of things (IoT) applications with varied requirements demand more flexibility and resilience from the cloud/edge computing architecture. In this article, to address the issues, we propose a novel air-ground integrated mobile edge network (AGMEN), where UAVs are flexibly deployed and scheduled, and assist the communication, caching, and computing of the edge network. In specific, we present the detailed architecture of AGMEN, and investigate the benefits and application scenarios of drone-cells, and UAV-assisted edge caching and computing. Furthermore, the challenging issues in AGMEN are discussed, and potential research directions are highlighted.

eess.SP↗