arXiv subjects

Bin Chen

Publications and source records attributed to Bin Chen.

At least 307 records · Page 17Linked to original sources

arXiv2024

Data-Aware Gradient Compression for FL in Communication-Constrained Mobile Computing

Federated Learning (FL) in mobile environments faces significant communication bottlenecks. Gradient compression has proven as an effective solution to this issue, offering substantial benefits in environments with limited bandwidth and metered data. Yet, it encounters severe performance drops in non-IID environments due to a one-size-fits-all compression approach, which does not account for the varying data volumes across workers. Assigning varying compression ratios to workers with distinct data distributions and volumes is therefore a promising solution. This work derives the convergence rate of distributed SGD with non-uniform compression, which reveals the intricate relationship between model convergence and the compression ratios applied to individual workers. Accordingly, we frame the relative compression ratio assignment as an $n$-variable chi-squared nonlinear optimization problem, constrained by a limited communication budget. We propose DAGC-R, which assigns conservative compression to workers handling larger data volumes. Recognizing the computational limitations of mobile devices, we propose the DAGC-A, which is computationally less demanding and enhances the robustness of compression in non-IID scenarios. Our experiments confirm that the DAGC-R and DAGC-A can speed up the training speed by up to $25.43\%$ and $16.65\%$ compared to the uniform compression respectively, when dealing with highly imbalanced data volume distribution and restricted communication.

cs.LG↗

arXiv2024

Practical Compact Deep Compressed Sensing

Recent years have witnessed the success of deep networks in compressed sensing (CS), which allows for a significant reduction in sampling cost and has gained growing attention since its inception. In this paper, we propose a new practical and compact network dubbed PCNet for general image CS. Specifically, in PCNet, a novel collaborative sampling operator is designed, which consists of a deep conditional filtering step and a dual-branch fast sampling step. The former learns an implicit representation of a linear transformation matrix into a few convolutions and first performs adaptive local filtering on the input image, while the latter then uses a discrete cosine transform and a scrambled block-diagonal Gaussian matrix to generate under-sampled measurements. Our PCNet is equipped with an enhanced proximal gradient descent algorithm-unrolled network for reconstruction. It offers flexibility, interpretability, and strong recovery performance for arbitrary sampling rates once trained. Additionally, we provide a deployment-oriented extraction scheme for single-pixel CS imaging systems, which allows for the convenient conversion of any linear sampling operator to its matrix form to be loaded onto hardware like digital micro-mirror devices. Extensive experiments on natural image CS, quantized CS, and self-supervised CS demonstrate the superior reconstruction accuracy and generalization ability of PCNet compared to existing state-of-the-art methods, particularly for high-resolution images. Code is available at https://github.com/Guaishou74851/PCNet.

cs.CV↗

arXiv2024

Leveraging NFTs for Spectrum Securitization in 6G Networks

Dynamic Spectrum Sharing can enhance spectrum resource utilization by promoting the dynamic distribution of spectrum resources. However, to effectively implement dynamic spectrum resource allocation, certain mechanisms are needed to incentivize primary users to proactively share their spectrum resources. This paper, based on the ERC404 standard and integrating Non-Fungible Token and Fungible Token technologies, proposes a spectrum securitization model to incentivize spectrum resource sharing and implements it on the Ethereum test net.

cs.CE↗

arXiv2024

QED Effects on Kerr-Newman Black Hole Shadows

Incorporating first-order QED effects, we explore the shadows of Kerr-Newman black holes with a magnetic charge through the numerical backward ray-tracing method. Our investigation accounts for both the direct influence of the electromagnetic field on light rays and the distortion of the background spacetime metric due to QED corrections. We notice that the area of the shadow increases with the QED effect, mainly due to the fact that the photons move more slowly in the effective medium and become easier to be trapped by the black hole.

gr-qc↗

arXiv2024

COSMIC: Compress Satellite Images Efficiently via Diffusion Compensation

With the rapidly increasing number of satellites in space and their enhanced capabilities, the amount of earth observation images collected by satellites is exceeding the transmission limits of satellite-to-ground links. Although existing learned image compression solutions achieve remarkable performance by using a sophisticated encoder to extract fruitful features as compression and using a decoder to reconstruct, it is still hard to directly deploy those complex encoders on current satellites' embedded GPUs with limited computing capability and power supply to compress images in orbit. In this paper, we propose COSMIC, a simple yet effective learned compression solution to transmit satellite images. We first design a lightweight encoder (i.e. reducing FLOPs by 2.6~5x) on satellite to achieve a high image compression ratio to save satellite-to-ground links. Then, for reconstructions on the ground, to deal with the feature extraction ability degradation due to simplifying encoders, we propose a diffusion-based model to compensate image details when decoding. Our insight is that satellite's earth observation photos are not just images but indeed multi-modal data with a nature of Text-to-Image pairing since they are collected with rich sensor data (e.g. coordinates, timestamp, etc.) that can be used as the condition for diffusion generation. Extensive experiments show that COSMIC outperforms state-of-the-art baselines on both perceptual and distortion metrics.

eess.IV↗

arXiv2024

Precession and Split of Tilted, Geometrically Thin Accretion Disk: an Analytical Study

It has been observed that many relativistic jets display a kind of cork-screw-like precession. Numerical simulations has suggested that such kind of precession may originate from the precession of the disk. In this work, we introduce an analytical model to describe the precession and split of a tilted, geometrically thin disk. We consider the Lense-Thirring effect from the central (primary) black hole (BH) and the gravitational effect from the companion (secondary) BH far away from the center, both of which could induce the precession of the accretion disk around the spin axis of central black hole. We propose the splitting conditions that when the rate of viscous diffusion cannot catch up with the dynamical frequency at a certain layer of fluid, the disk would split into two parts which precess independently. We presume that the precessions of the inner and outer disks are in accord with the rotation and precession of jet, respectively. By matching the frequencies of the disks to the observed frequencies of jet in the cork-screw-like precession and considering the splitting condition, we are allowed to read four parameters, the innermost radius ($r_{\rm in}$), the outermost radius ($r_{\rm out}$) of the disk, the initial splitting radius ($r_{\rm sp,0}$), and the inflow speed magnitude($β$), of the disk. We apply this model to OJ 287. Moreover, considering the inward shrinking of the disks, we find the time variation of the precession angle of jet. This time variation presents a unique feature of our model, which could be distinguishable in the future observation.

astro-ph.HE↗

arXiv2024

Energy Extraction from a Kerr Black Hole via Magnetic Reconnection within the Plunging Region

Magnetic reconnection within a highly magnetized plasma has been seen as a viable mechanism to extract the energy from a rotating black hole, as it can generate negative energy plasmoids in the ergoregion. For a typical accreting black hole, the ergoregion is filled with bulk plasma plunging from the innermost-stable-circular orbit (ISCO). In this study, we present an analytical study of the energy extraction via magnetic reconnection process in the plunging region. In contrast to the toroidal plasma, where the magnetic field cannot be derived from the MHD scheme, the magnetic field in the plunging plasma was determined by the ideal-MHD condition. We derive the global magnetic field structure in a fast reconnection model, and we read the expressions for the energies of plasmoids ejected from the reconnection region, for general stationary and axisymmetric spacetimes. Then, we demonstrate the behaviors of ejected energies varying with the reconnection locations in the Kerr spacetime, and identify the region where a negative-energy plasmoid can be produced. We find that for a certain magnetization there exists a critical value of the black hole spin, beyond which the energy extraction can occur, and the energy extraction is most efficient for the near-extreme black hole. Moreover, we study the conditions necessary for a plasmoid with positive energy to escape to the infinity, a crucial requirement for effective energy extractions. Considering the escaping conditions, we provide the parameter space in the radius-spin plane in which the energy extraction mechanism is effective.

gr-qc↗

arXiv2024

Energy extraction from a rotating black hole via magnetic reconnection: the plunging bulk plasma and orientation angle

Magnetic reconection provides a new avenue to extract energy from a rotating black hole in astrophysical scenarios. There have been many works studying the feasibility of extracting energy via magnetic reconnection in the scheme of magnetohydrodynamics. However, most of them focused on circularly flowing bulk plasma only, and the influence of orientation angle, the angle between flowing direction of bulk plasma and ejection direction of plasmoids, was never carefully considered. In this work, we would like to study the energy extraction via magnetic reconnection, via the so-called Comisso-Asenjo process specifically, in the plunging plasma. More importantly, we analyze the influence of orientation angle in depth. Actually, we consider the magnetic reconnection occurring in bulk plasma with two kinds of streamlines, one being the circular flow and the other being the plunging flow from the ISCO. We find that the plunging plasma has higher energy-extraction efficiency. Moreover, we notice that it is favorable for energy extraction when the orientation angle is suitably increased if the bulk plasma plunges. We further define the covering factor. Under the assumption that the magnetic reconnection occurs equally probable along any direction on any position within ergosphere, this value help us to primitively judge the probability of energy extraction via magnetic reconnection, through which we conclude that the plunging bulk plasma is much more capable in energy extraction than the circularly flowing one.

gr-qc↗

arXiv2024

Coport: A New Public Code for Polarized Radiative Transfer in a Covariant Framework$^\spadesuit$

General relativistic radiative transfer calculations are essential for comparing theoretical models of black hole accretion flows and jets with observational data. In this work, we introduce Coport, a novel public code specifically designed for covariant polarized ray-tracing radiative transfer computations in any spacetime. Written in Julia, Coport includes an interface for visualizing numerical results obtained from HARM, a publicly available implementation of the general relativistic magnetohydrodynamics code. We validate the precision of our code by comparing its outputs with the results from a variety of established methodologies. This includes the verification against analytical solutions, the validation through thin-disk assessments, and the evaluation via thick-disk analyses. Notably, our code employs a methodology that eliminates the need for separating the computations of spacetime propagation and plasma propagation. Instead, it directly solves the coupled, covariant, polarized radiative transfer equation in curved spacetime, seamlessly integrating the effects of gravity with plasma influences. This approach sets our code apart from the existing alternatives and enhances its accuracy and efficiency.

astro-ph.HE↗

arXiv2024

Geodesics in Carrollian Reissner-Nordström black holes

In this work, we study the geodesics in different types of Carrollian RN (Reissner-Nordström) black holes, considering the motions of both neutral and charged particles. We use the geodesic equations in the weak Carrollian structure and analyze the corresponding trajectories projected onto the absolute space, and find that the geodesics are well-defined. In particular, we examine the electric-electric and magnetic-electric limit of the RN black hole, focusing on their geodesic structures. We find that the global structures of the usual RN black holes get squeezed under the ultra-relativistic limit. More precisely, the nonextreme magnetic-electric RN spacetime has two different asymptotic flat patches while the extreme black hole spacetime consists of only one patch. For the magnetic-electric RN spacetime, the Carrollian extremal surfaces (CESs) divide the spacetime into several geodesically complete regions, and the geodesics can only travel in one of these regions. For the charged particles, we extend the analysis by considering their interactions with the electromagnetic field in the Carrollian RN spacetimes and find that their trajectories are significantly different from the neutral geodesics.

hep-th↗

arXiv2024

LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modeling

The pre-trained point cloud model based on Masked Point Modeling (MPM) has exhibited substantial improvements across various tasks. However, these models heavily rely on the Transformer, leading to quadratic complexity and limited decoder, hindering their practice application. To address this limitation, we first conduct a comprehensive analysis of existing Transformer-based MPM, emphasizing the idea that redundancy reduction is crucial for point cloud analysis. To this end, we propose a Locally constrained Compact point cloud Model (LCM) consisting of a locally constrained compact encoder and a locally constrained Mamba-based decoder. Our encoder replaces self-attention with our local aggregation layers to achieve an elegant balance between performance and efficiency. Considering the varying information density between masked and unmasked patches in the decoder inputs of MPM, we introduce a locally constrained Mamba-based decoder. This decoder ensures linear complexity while maximizing the perception of point cloud geometry information from unmasked patches with higher information density. Extensive experimental results show that our compact model significantly surpasses existing Transformer-based models in both performance and efficiency, especially our LCM-based Point-MAE model, compared to the Transformer-based model, achieved an improvement of 1.84%, 0.67%, and 0.60% in average accuracy on the three variants of ScanObjectNN while reducing parameters by 88% and computation by 73%. Code is available at https://github.com/zyh16143998882/LCM.

cs.CV↗

arXiv2024

Multi-path Exploration and Feedback Adjustment for Text-to-Image Person Retrieval

Text-based person retrieval aims to identify the specific persons using textual descriptions as queries. Existing ad vanced methods typically depend on vision-language pre trained (VLP) models to facilitate effective cross-modal alignment. However, the inherent constraints of VLP mod-els, which include the global alignment biases and insuffi-cient self-feedback regulation, impede optimal retrieval per formance. In this paper, we propose MeFa, a Multi-Pathway Exploration, Feedback, and Adjustment framework, which deeply explores intrinsic feedback of intra and inter-modal to make targeted adjustment, thereby achieving more precise person-text associations. Specifically, we first design an intra modal reasoning pathway that generates hard negative sam ples for cross-modal data, leveraging feedback from these samples to refine intra-modal reasoning, thereby enhancing sensitivity to subtle discrepancies. Subsequently, we intro duce a cross-modal refinement pathway that utilizes both global information and intermodal feedback to refine local in formation, thus enhancing its global semantic representation. Finally, the discriminative clue correction pathway incorpo rates fine-grained features of secondary similarity as discrim inative clues to further mitigate retrieval failures caused by disparities in these features. Experimental results on three public benchmarks demonstrate that MeFa achieves superior person retrieval performance without necessitating additional data or complex structures.

cs.CV↗

arXiv2024

BoostAdapter: Improving Vision-Language Test-Time Adaptation via Regional Bootstrapping

Adaptation of pretrained vision-language models such as CLIP to various downstream tasks have raised great interest in recent researches. Previous works have proposed a variety of test-time adaptation (TTA) methods to achieve strong generalization without any knowledge of the target domain. However, existing training-required TTA approaches like TPT necessitate entropy minimization that involves large computational overhead, while training-free methods like TDA overlook the potential for information mining from the test samples themselves. In this paper, we break down the design of existing popular training-required and training-free TTA methods and bridge the gap between them within our framework. Specifically, we maintain a light-weight key-value memory for feature retrieval from instance-agnostic historical samples and instance-aware boosting samples. The historical samples are filtered from the testing data stream and serve to extract useful information from the target distribution, while the boosting samples are drawn from regional bootstrapping and capture the knowledge of the test sample itself. We theoretically justify the rationality behind our method and empirically verify its effectiveness on both the out-of-distribution and the cross-domain datasets, showcasing its applicability in real-world situations.

cs.CV↗

arXiv2024

A Data-driven Crowd Simulation Framework Integrating Physics-informed Machine Learning with Navigation Potential Fields

Traditional rule-based physical models are limited by their reliance on singular physical formulas and parameters, making it difficult to effectively tackle the intricate tasks associated with crowd simulation. Recent research has introduced deep learning methods to tackle these issues, but most current approaches focus primarily on generating pedestrian trajectories, often lacking interpretability and failing to provide real-time dynamic simulations.To address the aforementioned issues, we propose a novel data-driven crowd simulation framework that integrates Physics-informed Machine Learning (PIML) with navigation potential fields. Our approach leverages the strengths of both physical models and PIML. Specifically, we design an innovative Physics-informed Spatio-temporal Graph Convolutional Network (PI-STGCN) as a data-driven module to predict pedestrian movement trends based on crowd spatio-temporal data. Additionally, we construct a physical model of navigation potential fields based on flow field theory to guide pedestrian movements, thereby reinforcing physical constraints during the simulation. In our framework, navigation potential fields are dynamically computed and updated based on the movement trends predicted by the PI-STGCN, while the updated crowd dynamics, guided by these fields, subsequently feed back into the PI-STGCN. Comparative experiments on two publicly available large-scale real-world datasets across five scenes demonstrate that our proposed framework outperforms existing rule-based methods in accuracy and fidelity. The similarity between simulated and actual pedestrian trajectories increases by 10.8%, while the average error is reduced by 4%. Moreover, our framework exhibits greater adaptability and better interpretability compared to methods that rely solely on deep learning for trajectory generation.

cs.AI↗

arXiv2024

Parameter Efficient Adaptation for Image Restoration with Heterogeneous Mixture-of-Experts

Designing single-task image restoration models for specific degradation has seen great success in recent years. To achieve generalized image restoration, all-in-one methods have recently been proposed and shown potential for multiple restoration tasks using one single model. Despite the promising results, the existing all-in-one paradigm still suffers from high computational costs as well as limited generalization on unseen degradations. In this work, we introduce an alternative solution to improve the generalization of image restoration models. Drawing inspiration from recent advancements in Parameter Efficient Transfer Learning (PETL), we aim to tune only a small number of parameters to adapt pre-trained restoration models to various tasks. However, current PETL methods fail to generalize across varied restoration tasks due to their homogeneous representation nature. To this end, we propose AdaptIR, a Mixture-of-Experts (MoE) with orthogonal multi-branch design to capture local spatial, global spatial, and channel representation bases, followed by adaptive base combination to obtain heterogeneous representation for different degradations. Extensive experiments demonstrate that our AdaptIR achieves stable performance on single-degradation tasks, and excels in hybrid-degradation tasks, with fine-tuning only 0.6% parameters for 8 hours.

cs.CV↗

arXiv2024

Towards Scalable Semantic Representation for Recommendation

With recent advances in large language models (LLMs), there has been emerging numbers of research in developing Semantic IDs based on LLMs to enhance the performance of recommendation systems. However, the dimension of these embeddings needs to match that of the ID embedding in recommendation, which is usually much smaller than the original length. Such dimension compression results in inevitable losses in discriminability and dimension robustness of the LLM embeddings, which motivates us to scale up the semantic representation. In this paper, we propose Mixture-of-Codes, which first constructs multiple independent codebooks for LLM representation in the indexing stage, and then utilizes the Semantic Representation along with a fusion module for the downstream recommendation stage. Extensive analysis and experiments demonstrate that our method achieves superior discriminability and dimension robustness scalability, leading to the best scale-up performance in recommendations.

cs.IR↗

arXiv2024

Editable-DeepSC: Cross-Modal Editable Semantic Communication Systems

Different from data-oriented communication systems that primarily focus on how to accurately transmit every bit of data, task-oriented semantic communication systems only transmit the specific semantic information required by downstream tasks, strive to minimize the communication overhead and maintain competitive tasks execution performance in the presence of channel noise. However, it is worth noting that in many scenarios, the transmitted semantic information needs to be dynamically modified according to the users' preferences in a conversational and interactive way, which few existing works take into consideration. In this paper, we propose a novel cross-modal editable semantic communication system, named Editable-DeepSC, to tackle this challenge. By utilizing inversion methods based on StyleGAN priors, Editable-DeepSC takes cross-modal text-image pairs as the inputs and transmits the edited information of images based on textual instructions. Extensive numerical results demonstrate that our proposed Editable-DeepSC can achieve remarkable editing effects and transmission efficiency under the perturbations of channel noise, outperforming existing data-oriented communication methods.

cs.IT↗

arXiv2024

ReFIR: Grounding Large Restoration Models with Retrieval Augmentation

Recent advances in diffusion-based Large Restoration Models (LRMs) have significantly improved photo-realistic image restoration by leveraging the internal knowledge embedded within model weights. However, existing LRMs often suffer from the hallucination dilemma, i.e., producing incorrect contents or textures when dealing with severe degradations, due to their heavy reliance on limited internal knowledge. In this paper, we propose an orthogonal solution called the Retrieval-augmented Framework for Image Restoration (ReFIR), which incorporates retrieved images as external knowledge to extend the knowledge boundary of existing LRMs in generating details faithful to the original scene. Specifically, we first introduce the nearest neighbor lookup to retrieve content-relevant high-quality images as reference, after which we propose the cross-image injection to modify existing LRMs to utilize high-quality textures from retrieved images. Thanks to the additional external knowledge, our ReFIR can well handle the hallucination challenge and facilitate faithfully results. Extensive experiments demonstrate that ReFIR can achieve not only high-fidelity but also realistic restoration results. Importantly, our ReFIR requires no training and is adaptable to various LRMs.

cs.CV↗