arXiv subjects

Xiaofeng Wu

Publications and source records attributed to Xiaofeng Wu.

At least 37 records · Page 2Linked to original sources

arXiv2025

The Impact Analysis of Delays in Asynchronous Federated Learning with Data Heterogeneity for Edge Intelligence

Federated learning (FL) has provided a new methodology for coordinating a group of clients to train a machine learning model collaboratively, bringing an efficient paradigm in edge intelligence. Despite its promise, FL faces several critical challenges in practical applications involving edge devices, such as data heterogeneity and delays stemming from communication and computation constraints. This paper examines the impact of unknown causes of delay on training performance in an Asynchronous Federated Learning (AFL) system with data heterogeneity. Initially, an asynchronous error definition is proposed, based on which the solely adverse impact of data heterogeneity is theoretically analyzed within the traditional Synchronous Federated Learning (SFL) framework. Furthermore, Asynchronous Updates with Delayed Gradients (AUDG), a conventional AFL scheme, is discussed. Investigation into AUDG reveals that the negative influence of data heterogeneity is correlated with delays, while a shorter average delay from a specific client does not consistently enhance training performance. In order to compensate for the scenarios where AUDG are not adapted, Pseudo-synchronous Updates by Reusing Delayed Gradients (PSURDG) is proposed, and its theoretical convergence is analyzed. In both AUDG and PSURDG, only a random set of clients successfully transmits their updated results to the central server in each iteration. The critical difference between them lies in whether the delayed information is reused. Finally, both schemes are validated and compared through theoretical analysis and simulations, demonstrating more intuitively that discarding outdated information due to time delays is not always the best approach.

cs.LG↗

arXiv2025

Temporal Misalignment in ANN-SNN Conversion and Its Mitigation via Probabilistic Spiking Neurons

Spiking Neural Networks (SNNs) offer a more energy-efficient alternative to Artificial Neural Networks (ANNs) by mimicking biological neural principles, establishing them as a promising approach to mitigate the increasing energy demands of large-scale neural models. However, fully harnessing the capabilities of SNNs remains challenging due to their discrete signal processing and temporal dynamics. ANN-SNN conversion has emerged as a practical approach, enabling SNNs to achieve competitive performance on complex machine learning tasks. In this work, we identify a phenomenon in the ANN-SNN conversion framework, termed temporal misalignment, in which random spike rearrangement across SNN layers leads to performance improvements. Based on this observation, we introduce biologically plausible two-phase probabilistic (TPP) spiking neurons, further enhancing the conversion process. We demonstrate the advantages of our proposed method both theoretically and empirically through comprehensive experiments on CIFAR-10/100, CIFAR10-DVS, and ImageNet across a variety of architectures, achieving state-of-the-art results.

cs.LG↗

arXiv2025

The Impact of Visual Information in Chinese Characters: Evaluating Large Models' Ability to Recognize and Utilize Radicals

The glyphic writing system of Chinese incorporates information-rich visual features in each character, such as radicals that provide hints about meaning or pronunciation. However, there has been no investigation into whether contemporary Large Language Models (LLMs) and Vision-Language Models (VLMs) can harness these sub-character features in Chinese through prompting. In this study, we establish a benchmark to evaluate LLMs' and VLMs' understanding of visual elements in Chinese characters, including radicals, composition structures, strokes, and stroke counts. Our results reveal that models surprisingly exhibit some, but still limited, knowledge of the visual information, regardless of whether images of characters are provided. To incite models' ability to use radicals, we further experiment with incorporating radicals into the prompts for Chinese language processing (CLP) tasks. We observe consistent improvement in Part-Of-Speech tagging when providing additional information about radicals, suggesting the potential to enhance CLP by integrating sub-character information.

cs.CL↗

arXiv2025

UAV-Assisted Multi-Task Federated Learning with Task Knowledge Sharing

The rapid development of Unmanned aerial vehicles (UAVs) technology has spawned a wide variety of applications, such as emergency communications, regional surveillance, and disaster relief. Due to their limited battery capacity and processing power, multiple UAVs are often required for complex tasks. In such cases, a control center is crucial for coordinating their activities, which fits well with the federated learning (FL) framework. However, conventional FL approaches often focus on a single task, ignoring the potential of training multiple related tasks simultaneously. In this paper, we propose a UAV-assisted multi-task federated learning scheme, in which data collected by multiple UAVs can be used to train multiple related tasks concurrently. The scheme facilitates the training process by sharing feature extractors across related tasks and introduces a task attention mechanism to balance task performance and encourage knowledge sharing. To provide an analytical description of training performance, the convergence analysis of the proposed scheme is performed. Additionally, the optimal bandwidth allocation for UAVs under limited bandwidth conditions is derived to minimize communication time. Meanwhile, a UAV-EV association strategy based on coalition formation game is proposed. Simulation results validate the effectiveness of the proposed scheme in enhancing multi-task performance and training speed.

cs.LG↗

arXiv2024

GSD: View-Guided Gaussian Splatting Diffusion for 3D Reconstruction

We present GSD, a diffusion model approach based on Gaussian Splatting (GS) representation for 3D object reconstruction from a single view. Prior works suffer from inconsistent 3D geometry or mediocre rendering quality due to improper representations. We take a step towards resolving these shortcomings by utilizing the recent state-of-the-art 3D explicit representation, Gaussian Splatting, and an unconditional diffusion model. This model learns to generate 3D objects represented by sets of GS ellipsoids. With these strong generative 3D priors, though learning unconditionally, the diffusion model is ready for view-guided reconstruction without further model fine-tuning. This is achieved by propagating fine-grained 2D features through the efficient yet flexible splatting function and the guided denoising sampling process. In addition, a 2D diffusion model is further employed to enhance rendering fidelity, and improve reconstructed GS quality by polishing and re-using the rendered images. The final reconstructed objects explicitly come with high-quality 3D structure and texture, and can be efficiently rendered in arbitrary views. Experiments on the challenging real-world CO3D dataset demonstrate the superiority of our approach. Project page: https://yxmu.foo/GSD/

cs.CV↗

arXiv2024

FTBC: Forward Temporal Bias Correction for Optimizing ANN-SNN Conversion

Spiking Neural Networks (SNNs) offer a promising avenue for energy-efficient computing compared with Artificial Neural Networks (ANNs), closely mirroring biological neural processes. However, this potential comes with inherent challenges in directly training SNNs through spatio-temporal backpropagation -- stemming from the temporal dynamics of spiking neurons and their discrete signal processing -- which necessitates alternative ways of training, most notably through ANN-SNN conversion. In this work, we introduce a lightweight Forward Temporal Bias Correction (FTBC) technique, aimed at enhancing conversion accuracy without the computational overhead. We ground our method on provided theoretical findings that through proper temporal bias calibration the expected error of ANN-SNN conversion can be reduced to be zero after each time step. We further propose a heuristic algorithm for finding the temporal bias only in the forward pass, thus eliminating the computational burden of backpropagation and we evaluate our method on CIFAR-10/100 and ImageNet datasets, achieving a notable increase in accuracy on all datasets. Codes are released at a GitHub repository.

cs.AI↗

arXiv2024

ATOM: Asynchronous Training of Massive Models for Deep Learning in a Decentralized Environment

The advent of the Transformer architecture has propelled the growth of natural language processing (NLP) models, leading to remarkable achievements in numerous NLP tasks. Yet, the absence of specialized hardware like expansive GPU memory and high-speed interconnects poses challenges for training large-scale models. This makes it daunting for many users to experiment with pre-training and fine-tuning large language models (LLMs). In this study, we introduce \atom, a resilient distributed training framework designed for asynchronous training of vast models in a decentralized setting using cost-effective hardware, including consumer-grade GPUs and Ethernet. Unlike conventional model partitioning methods that distribute sub-models across GPUs, \atom aims to accommodate a complete LLM on one host (peer) through seamlessly model swapping and concurrently trains multiple copies across various peers to optimize training throughput. Through static analysis, \atom identifies the best model partitioning strategy and flawlessly merges model execution with swapping. Key benefits of \atom include: Avoiding the central point of failure found in pipeline parallelism methods. Demonstrating superior performance and scalability compared to closely-integrated pipeline parallelism in slower networks. Our experiments using different GPT-3 model configurations reveal that, in scenarios with suboptimal network connections, \atom can enhance training efficiency up to $20 \times$ when juxtaposed with the state-of-the-art decentralized pipeline parallelism approaches.

cs.DC↗

arXiv2024

Indoor Positioning based on Active Radar Sensing and Passive Reflectors: Concepts & Initial Results

To navigate reliably in indoor environments, an industrial autonomous vehicle must know its position. However, current indoor vehicle positioning technologies either lack accuracy, usability or are too expensive. Thus, we propose a novel concept called local reference point assisted active radar positioning, which is able to overcome these drawbacks. It is based on distributing passive retroreflectors in the indoor environment such that each position of the vehicle can be identified by a unique reflection characteristic regarding the reflectors. To observe these characteristics, the autonomous vehicle is equipped with an active radar system. On one hand, this paper presents the basic idea and concept of our new approach towards indoor vehicle positioning and especially focuses on the crucial placement of the reflectors. On the other hand, it also provides a proof of concept by conducting a full system simulation including the placement of the local reference points, the radar-based distance estimation and the comparison of two different positioning methods. It successfully demonstrates the feasibility of our proposed approach.

cs.RO↗

arXiv2024

Weak antilocalization and localization in Eu$_2$Ir$_2$O$_7$ (111) thin films by reactive solid phase epitaxy

Thin films of the pyrochlore iridates along the [111] direction have drawn significant attention to investigate exotic correlated topological phenomena. Here, we report the fabrication of Eu$_2$Ir$_2$O$_7$ thin films via reactive solid phase epitaxy using the pulsed laser deposition technique. We mainly focus on the transport properties of the films below the magnetic phase transition at 105 K. Analyses on the temperature and the field dependences of resistivity unveil the presence of weak antilocalization, a characteristic signature of the Weyl semimetallic state that has been "buried" by magnetism. Moreover, it is noteworthy that the contribution from many-body interactions in Eu2Ir2O7 thin films is enhanced at lower temperatures and competes with the weak antilocalization effect, and eventually drives the crossover to weak localization at 2 K.

cond-mat.str-el↗

arXiv2023

Magnetism and berry phase manipulation in an emergent structure of perovskite ruthenate by (111) strain engineering

The interplay among symmetry of lattices, electronic correlations, and Berry phase of the Bloch states in solids has led to fascinating quantum phases of matter. A prototypical system is the magnetic Weyl candidate SrRuO3, where designing and creating electronic and topological properties on artificial lattice geometry is highly demanded yet remains elusive. Here, we establish an emergent trigonal structure of SrRuO3 by means of heteroepitaxial strain engineering along the [111] crystallographic axis. Distinctive from bulk, the trigonal SrRuO3 exhibits a peculiar XY-type ferromagnetic ground state, with the coexistence of high-mobility holes likely from linear Weyl bands and low-mobility electrons from normal quadratic bands as carriers. The presence of Weyl nodes are further corroborated by capturing intrinsic anomalous Hall effect, acting as momentum-space sources of Berry curvatures. The experimental observations are consistent with our first-principles calculations, shedding light on the detailed band topology of trigonal SrRuO3 with multiple pairs of Weyl nodes near the Fermi level. Our findings signify the essence of magnetism and Berry phase manipulation via lattice design and pave the way towards unveiling nontrivial correlated topological phenomena.

cond-mat.str-el↗

arXiv2023

A Simulation-Augmented Benchmarking Framework for Automatic RSO Streak Detection in Single-Frame Space Images

Detecting Resident Space Objects (RSOs) and preventing collisions with other satellites is crucial. Recently, deep convolutional neural networks (DCNNs) have shown superior performance in object detection when large-scale datasets are available. However, collecting rich data of RSOs is difficult due to very few occurrences in the space images. Without sufficient data, it is challenging to comprehensively train DCNN detectors and make them effective for detecting RSOs in space images, let alone to estimate whether a detector is sufficiently robust. The lack of meaningful evaluation of different detectors could further affect the design and application of detection methods. To tackle this issue, we propose that the space images containing RSOs can be simulated to complement the shortage of raw data for better benchmarking. Accordingly, we introduce a novel simulation-augmented benchmarking framework for RSO detection (SAB-RSOD). In our framework, by making the best use of the hardware parameters of the sensor that captures real-world space images, we first develop a high-fidelity RSO simulator that can generate various realistic space images. Then, we use this simulator to generate images that contain diversified RSOs in space and annotate them automatically. Later, we mix the synthetic images with the real-world images, obtaining around 500 images for training with only the real-world images for evaluation. Under SAB-RSOD, we can train different popular object detectors like Yolo and Faster RCNN effectively, enabling us to evaluate their performance thoroughly. The evaluation results have shown that the amount of available data and image resolution are two key factors for robust RSO detection. Moreover, if using a lower resolution for higher efficiency, we demonstrated that a simple UNet-based detection method can already access high detection accuracy.

cs.CV↗

arXiv2023

Doppler-Resilient Universal Filtered MultiCarrier (DR-UFMC): A Beyond-OTFS Modulation

In the past few years, some alternatives to the Orthogonal Frequency Division Multiplexing (OFDM) modulation have been considered to improve its spectral containment and its performance level in the presence of heavy Doppler shifts. This paper examines a novel modulation, named Doppler-Resilient Universal Filtered MultiCarrier (DR-UFMC), which has the objective of combining the advantages provided by the Universal Filtered MultiCarrier (UFMC) modulation (i.e., better spectral containment), with those of the Orthogonal Time Frequency Space (OTFS) modulation (i.e., better performance in time-varying environments). The paper contains the mathematical model and detailed transceiver block scheme of the newly described modulation, along with a numerical analysis contrasting DR-UFMC against OTFS, OFDM with one-tap frequency domain equalization (FDE), and OFDM with multicarrier multisymbol linear MMSE processing. Results clearly show the superiority, with respect to the cited benchmarks, of the newly proposed modulation in terms of achievable spectral efficiency. Interestingly, it is also seen that OFDM, when considered in conjunction with multicarrier multisymbol linear minimum mean squares error (MMSE) processing, performs slightly better than OTFS in terms of achievable spectral efficiency.

cs.IT↗

arXiv2023

Efficient 3D Object Reconstruction using Visual Transformers

Reconstructing a 3D object from a 2D image is a well-researched vision problem, with many kinds of deep learning techniques having been tried. Most commonly, 3D convolutional approaches are used, though previous work has shown state-of-the-art methods using 2D convolutions that are also significantly more efficient to train. With the recent rise of transformers for vision tasks, often outperforming convolutional methods, along with some earlier attempts to use transformers for 3D object reconstruction, we set out to use visual transformers in place of convolutions in existing efficient, high-performing techniques for 3D object reconstruction in order to achieve superior results on the task. Using a transformer-based encoder and decoder to predict 3D structure from 2D images, we achieve accuracy similar or superior to the baseline approach. This study serves as evidence for the potential of visual transformers in the task of 3D object reconstruction.

cs.CV↗

arXiv2022

An Optimised Satellite Constellation for Forest Fire Detection through Edge Computing

The end of 2019 marked a bushfire crisis for Australia that affected more than 100000km2 of land and destroyed more than 2000 houses. Here, we propose a method of in-orbit bushfire detection with high efficiency to prevent a repetition of this disaster. An LEO satellite constellation is first developed through NSGA-II (Nondominated Sorting Genetic Algorithm II), optimising for coverage over Australia. Then edge computing is adopted to run a bushfire detection algorithm using several constellation satellites as edge nodes to reduce fire detection time. A geostationary satellite is used for inter-satellite communications, such that an image taken by a satellite can be distributed among several satellites for processing. The geostationary satellite also maintains a constant link to the ground, so that a bushfire detection can be reported back without any significant delay. Overall, this system is able to detect fires that span more than 5m in length, and can make detections in 1.39s per image processed. This is faster than any currently available bushfire detection method.

math.OC↗

arXiv2022

Multi-UE Multi-AP Beam Alignment in User-Centric Cell-Free Massive MIMO Systems Operating at mmWave

This paper considers the problem of beam alignment in a cell-free massive MIMO deployment with multiple access points (APs) and multiple user equipments (UEs) simultaneously operating in the same millimeter wave frequency band. Assuming the availability of a control channel at sub-6 GHz frequencies, a protocol is developed that permits estimating, for each UE, the strongest propagation path from each of the surrounding APs, and to perform user-centric association between the UEs and the APs. Estimation of the strongest paths from nearby APs is realized at the UE in a one-phase procedure, during which all the APs simultaneously transmit on pseudo-randomly selected channels with pseudo-random transmit beamformers. An algorithm for orthogonal channels assignment to the APs is also proposed, with the aim of minimizing the mutual interference between APs that transmit on the same channels. The performance of the proposed strategy is evaluated both in terms of probability of correct detection of the directions of arrival and of departure associated to the strongest beam from nearby APs, and in terms of downlink and uplink signal-to-interference-plus-noise ratio. Numerical results show that the proposed approach is effective and capable of efficiently realizing beam alignment in a multi-UE multi-AP wireless scenario.

cs.IT↗

arXiv2021

Densely Semantic Enhancement for Domain Adaptive Region-free Detectors

Unsupervised domain adaptive object detection aims to adapt a well-trained detector from its original source domain with rich labeled data to a new target domain with unlabeled data. Previous works focus on improving the domain adaptability of region-based detectors, e.g., Faster-RCNN, through matching cross-domain instance-level features that are explicitly extracted from a region proposal network (RPN). However, this is unsuitable for region-free detectors such as single shot detector (SSD), which perform a dense prediction from all possible locations in an image and do not have the RPN to encode such instance-level features. As a result, they fail to align important image regions and crucial instance-level features between the domains of region-free detectors. In this work, we propose an adversarial module to strengthen the cross-domain matching of instance-level features for region-free detectors. Firstly, to emphasize the important regions of image, the DSEM learns to predict a transferable foreground enhancement mask that can be utilized to suppress the background disturbance in an image. Secondly, considering that region-free detectors recognize objects of different scales using multi-scale feature maps, the DSEM encodes both multi-level semantic representations and multi-instance spatial-contextual relationships across different domains. Finally, the DSEM is pluggable into different region-free detectors, ultimately achieving the densely semantic feature matching via adversarial learning. Extensive experiments have been conducted on PASCAL VOC, Clipart, Comic, Watercolor, and FoggyCityscape benchmarks, and their results well demonstrate that the proposed approach not only improves the domain adaptability of region-free detectors but also outperforms existing domain adaptive region-based detectors under various domain shift settings.

cs.CV↗

arXiv2021

Beam Alignment in mmWave User-Centric Cell-Free Massive MIMO Systems

The problem of beam alignment (BA) in a cell-free massive multiple-input multiple-output (CF-mMIMO) system operating at millimeter wave (mmWaves) carrier frequencies is considered in this paper. Two estimation algorithms are proposed, in association with a protocol that permits simultaneous estimation, on a shared set of frequencies, for each user equipment (UE), of the direction of arrival and departure of the radio waves associated to the strongest propagation paths from each of the surrounding access points (APs), so that UE-AP association can take place. The proposed procedure relies on the existence of a reliable control channel at sub-6 GHz frequency, so as to enable exchange of estimated values between the UEs and the network, and assumes that APs can be identifies based on the prior knowledge of the orthogonal channels and transmit beamforming codebook. A strategy for assigning codebook entries to the several APs is also proposed, with the aim of minimizing the mutual interference between APs that are assigned the same entry. Numerical results show the effectiveness of the proposed detection strategy, thus enabling one shot fast BA for CF-mMIMO systems.

cs.IT↗

arXiv2019

A Model-driven and Data-driven Fusion Framework for Accurate Air Quality Prediction

Air quality is closely related to public health. Health issues such as cardiovascular diseases and respiratory diseases, may have connection with long exposure to highly polluted environment. Therefore, accurate air quality forecasts are extremely important to those who are vulnerable. To estimate the variation of several air pollution concentrations, previous researchers used various approaches, such as the Community Multiscale Air Quality model (CMAQ) or neural networks. Although CMAQ model considers a coverage of the historic air pollution data and meteorological variables, extra bias is introduced due to additional adjustment. In this paper, a combination of model-based strategy and data-driven method namely the physical-temporal collection(PTC) model is proposed, aiming to fix the systematic error that traditional models deliver. In the data-driven part, the first components are the temporal pattern and the weather pattern to measure important features that contribute to the prediction performance. The less relevant input variables will be removed to eliminate negative weights in network training. Then, we deploy a long-short-term-memory (LSTM) to fetch the preliminary results, which will be further corrected by a neural network (NN) involving the meteorological index as well as other pollutants concentrations. The data-set we applied for forecasting is from January 1st, 2016 to December 31st, 2016. According to the results, our PTC achieves an excellent performance compared with the baseline model (CMAQ prediction, GRU, DNN and etc.). This joint model-based data-driven method for air quality prediction can be easily deployed on stations without extra adjustment, providing results with high-time-resolution information for vulnerable members to prevent heavy air pollution ahead.

stat.AP↗