arXiv subjects

Tian Wang

Publications and source records attributed to Tian Wang.

At least 73 records · Page 4Linked to original sources

arXiv2025

LRScheduler: A Layer-aware and Resource-adaptive Container Scheduler in Edge Computing

Lightweight containers provide an efficient approach for deploying computation-intensive applications in network edge. The layered storage structure of container images can further reduce the deployment cost and container startup time. Existing researches discuss layer sharing scheduling theoretically but with little attention paid to the practical implementation. To fill in this gap, we propose and implement a Layer-aware and Resource-adaptive container Scheduler (LRScheduler) in edge computing. Specifically, we first utilize container image layer information to design and implement a node scoring and container scheduling mechanism. This mechanism can effectively reduce the download cost when deploying containers, which is very important in edge computing with limited bandwidth. Then, we design a dynamically weighted and resource-adaptive mechanism to enhance load balancing in edge clusters, increasing layer sharing scores when resource load is low to use idle resources effectively. Our scheduler is built on the scheduling framework of Kubernetes, enabling full process automation from task information acquisition to container dep=loyment. Testing on a real system has shown that our design can effectively reduce the container deployment cost as compared with the default scheduler.

cs.DC↗

arXiv2025

Hybrid Learning for Cold-Start-Aware Microservice Scheduling in Dynamic Edge Environments

With the rapid growth of IoT devices and their diverse workloads, container-based microservices deployed at edge nodes have become a lightweight and scalable solution. However, existing microservice scheduling algorithms often assume static resource availability, which is unrealistic when multiple containers are assigned to an edge node. Besides, containers suffer from cold-start inefficiencies during early-stage training in currently popular reinforcement learning (RL) algorithms. In this paper, we propose a hybrid learning framework that combines offline imitation learning (IL) with online Soft Actor-Critic (SAC) optimization to enable a cold-start-aware microservice scheduling with dynamic allocation for computing resources. We first formulate a delay-and-energy-aware scheduling problem and construct a rule-based expert to generate demonstration data for behavior cloning. Then, a GRU-enhanced policy network is designed in the policy network to extract the correlation among multiple decisions by separately encoding slow-evolving node states and fast-changing microservice features, and an action selection mechanism is given to speed up the convergence. Extensive experiments show that our method significantly accelerates convergence and achieves superior final performance. Compared with baselines, our algorithm improves the total objective by $50\%$ and convergence speed by $70\%$, and demonstrates the highest stability and robustness across various edge configurations.

cs.NI↗

arXiv2025

FedHL: Federated Learning for Heterogeneous Low-Rank Adaptation via Unbiased Aggregation

Federated Learning (FL) facilitates the fine-tuning of Foundation Models (FMs) using distributed data sources, with Low-Rank Adaptation (LoRA) gaining popularity due to its low communication costs and strong performance. While recent work acknowledges the benefits of heterogeneous LoRA in FL and introduces flexible algorithms to support its implementation, our theoretical analysis reveals a critical gap: existing methods lack formal convergence guarantees due to parameter truncation and biased gradient updates. Specifically, adapting client-specific LoRA ranks necessitates truncating global parameters, which introduces inherent truncation errors and leads to subsequent inaccurate gradient updates that accumulate over training rounds, ultimately degrading performance. To address the above issues, we propose \textbf{FedHL}, a simple yet effective \textbf{Fed}erated Learning framework tailored for \textbf{H}eterogeneous \textbf{L}oRA. By leveraging the full-rank global model as a calibrated aggregation basis, FedHL eliminates the direct truncation bias from initial alignment with client-specific ranks. Furthermore, we derive the theoretically optimal aggregation weights by minimizing the gradient drift term in the convergence upper bound. Our analysis shows that FedHL guarantees $\mathcal{O}(1/\sqrt{T})$ convergence rate, and experiments on multiple real-world datasets demonstrate a 1-3\% improvement over several state-of-the-art methods.

cs.LG↗

arXiv2025

AFCL: Analytic Federated Continual Learning for Spatio-Temporal Invariance of Non-IID Data

Federated Continual Learning (FCL) enables distributed clients to collaboratively train a global model from online task streams in dynamic real-world scenarios. However, existing FCL methods face challenges of both spatial data heterogeneity among distributed clients and temporal data heterogeneity across online tasks. Such data heterogeneity significantly degrades the model performance with severe spatial-temporal catastrophic forgetting of local and past knowledge. In this paper, we identify that the root cause of this issue lies in the inherent vulnerability and sensitivity of gradients to non-IID data. To fundamentally address this issue, we propose a gradient-free method, named Analytic Federated Continual Learning (AFCL), by deriving analytical (i.e., closed-form) solutions from frozen extracted features. In local training, our AFCL enables single-epoch learning with only a lightweight forward-propagation process for each client. In global aggregation, the server can recursively and efficiently update the global model with single-round aggregation. Theoretical analyses validate that our AFCL achieves spatio-temporal invariance of non-IID data. This ideal property implies that, regardless of how heterogeneous the data are distributed across local clients and online tasks, the aggregated model of our AFCL remains invariant and identical to that of centralized joint learning. Extensive experiments show the consistent superiority of our AFCL over state-of-the-art baselines across various benchmark datasets and settings.

cs.LG↗

arXiv2025

InfoPO: On Mutual Information Maximization for Large Language Model Alignment

We study the post-training of large language models (LLMs) with human preference data. Recently, direct preference optimization and its variants have shown considerable promise in aligning language models, eliminating the need for reward models and online sampling. Despite these benefits, these methods rely on explicit assumptions about the Bradley-Terry (BT) model, which makes them prone to overfitting and results in suboptimal performance, particularly on reasoning-heavy tasks. To address these challenges, we propose a principled preference fine-tuning algorithm called InfoPO, which effectively and efficiently aligns large language models using preference data. InfoPO eliminates the reliance on the BT model and prevents the likelihood of the chosen response from decreasing. Extensive experiments confirm that InfoPO consistently outperforms established baselines on widely used open benchmarks, particularly in reasoning tasks.

cs.LG↗

arXiv2025

RDPP-TD: Reputation and Data Privacy-Preserving based Truth Discovery Scheme in Mobile Crowdsensing

Truth discovery (TD) plays an important role in Mobile Crowdsensing (MCS). However, existing TD methods, including privacy-preserving TD approaches, estimate the truth by weighting only the data submitted in the current round, which often results in low data quality. Moreover, there is a lack of effective TD methods that preserve both reputation and data privacy. To address these issues, a Reputation and Data Privacy-Preserving based Truth Discovery (RDPP-TD) scheme is proposed to obtain high-quality data for MCS. The RDPP-TD scheme consists of two key approaches: a Reputation-based Truth Discovery (RTD) approach, which integrates the weight of current-round data with workers' reputation values to estimate the truth, thereby achieving more accurate results, and a Reputation and Data Privacy-Preserving (RDPP) approach, which ensures privacy preservation for sensing data and reputation values. First, the RDPP approach, when seamlessly integrated with RTD, can effectively evaluate the reliability of workers and their sensing data in a privacy-preserving manner. Second, the RDPP scheme supports reputation-based worker recruitment and rewards, ensuring high-quality data collection while incentivizing workers to provide accurate information. Comprehensive theoretical analysis and extensive experiments based on real-world datasets demonstrate that the proposed RDPP-TD scheme provides strong privacy protection and improves data quality by up to 33.3%.

cs.CE↗

arXiv2025

Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models

The rapid advancement of artificial intelligence (AI) technologies has led to an increasing deployment of AI models on edge and terminal devices, driven by the proliferation of the Internet of Things (IoT) and the need for real-time data processing. This survey comprehensively explores the current state, technical challenges, and future trends of on-device AI models. We define on-device AI models as those designed to perform local data processing and inference, emphasizing their characteristics such as real-time performance, resource constraints, and enhanced data privacy. The survey is structured around key themes, including the fundamental concepts of AI models, application scenarios across various domains, and the technical challenges faced in edge environments. We also discuss optimization and implementation strategies, such as data preprocessing, model compression, and hardware acceleration, which are essential for effective deployment. Furthermore, we examine the impact of emerging technologies, including edge computing and foundation models, on the evolution of on-device AI models. By providing a structured overview of the challenges, solutions, and future directions, this survey aims to facilitate further research and application of on-device AI, ultimately contributing to the advancement of intelligent systems in everyday life.

cs.AI↗

arXiv2025

Isogenies of CM Elliptic Curves

Given two CM elliptic curves over a number field and a natural number $m$, we establish a polynomial lower bound (in terms of $m$) for the number of rational primes $p$ such that the reductions of these elliptic curves modulo a prime above $p$ are $m$-isogenous. The proof relies on higher Green functions and theorems of Gross-Zagier and Gross-Kohnen-Zagier. A crucial observation is that the Fourier coefficients of incoherent Eisenstein series can be approximated by those of coherent Eisenstein series of increasing level. Another key ingredient is an explicit upper bound for the Petersson norm of an arbitrary elliptic modular form in terms of finitely many of its Fourier coefficients at the cusp infinity, which is a result of independent interest.

math.NT↗

arXiv2025

Unveiling Privacy and Security Gaps in Female Health Apps

Female Health Applications (FHA), a growing segment of FemTech, aim to provide affordable and accessible healthcare solutions for women globally. These applications gather and monitor health and reproductive data from millions of users. With ongoing debates on women's reproductive rights and privacy, it's crucial to assess how these apps protect users' privacy. In this paper, we undertake a security and data protection assessment of 45 popular FHAs. Our investigation uncovers harmful permissions, extensive collection of sensitive personal and medical data, and the presence of numerous third-party tracking libraries. Furthermore, our examination of their privacy policies reveals deviations from fundamental data privacy principles. These findings highlight a significant lack of privacy and security measures for FemTech apps, especially as women's reproductive rights face growing political challenges. The results and recommendations provide valuable insights for users, app developers, and policymakers, paving the way for better privacy and security in Female Health Applications.

cs.HC↗

arXiv2024

STNMamba: Mamba-based Spatial-Temporal Normality Learning for Video Anomaly Detection

Video anomaly detection (VAD) has been extensively researched due to its potential for intelligent video systems. However, most existing methods based on CNNs and transformers still suffer from substantial computational burdens and have room for improvement in learning spatial-temporal normality. Recently, Mamba has shown great potential for modeling long-range dependencies with linear complexity, providing an effective solution to the above dilemma. To this end, we propose a lightweight and effective Mamba-based network named STNMamba, which incorporates carefully designed Mamba modules to enhance the learning of spatial-temporal normality. Firstly, we develop a dual-encoder architecture, where the spatial encoder equipped with Multi-Scale Vision Space State Blocks (MS-VSSB) extracts multi-scale appearance features, and the temporal encoder employs Channel-Aware Vision Space State Blocks (CA-VSSB) to capture significant motion patterns. Secondly, a Spatial-Temporal Interaction Module (STIM) is introduced to integrate spatial and temporal information across multiple levels, enabling effective modeling of intrinsic spatial-temporal consistency. Within this module, the Spatial-Temporal Fusion Block (STFB) is proposed to fuse the spatial and temporal features into a unified feature space, and the memory bank is utilized to store spatial-temporal prototypes of normal patterns, restricting the model's ability to represent anomalies. Extensive experiments on three benchmark datasets demonstrate that our STNMamba achieves competitive performance with fewer parameters and lower computational costs than existing methods.

cs.CV↗

arXiv2024

Accelerating AIGC Services with Latent Action Diffusion Scheduling in Edge Networks

Artificial Intelligence Generated Content (AIGC) has gained significant popularity for creating diverse content. Current AIGC models primarily focus on content quality within a centralized framework, resulting in a high service delay and negative user experiences. However, not only does the workload of an AIGC task depend on the AIGC model's complexity rather than the amount of data, but the large model and its multi-layer encoder structure also result in a huge demand for computational and memory resources. These unique characteristics pose new challenges in its modeling, deployment, and scheduling at edge networks. Thus, we model an offloading problem among edges for providing real AIGC services and propose LAD-TS, a novel Latent Action Diffusion-based Task Scheduling method that orchestrates multiple edge servers for expedited AIGC services. The LAD-TS generates a near-optimal offloading decision by leveraging the diffusion model's conditional generation capability and the reinforcement learning's environment interaction ability, thereby minimizing the service delays under multiple resource constraints. Meanwhile, a latent action diffusion strategy is designed to guide decision generation by utilizing historical action probability, enabling rapid achievement of near-optimal decisions. Furthermore, we develop DEdgeAI, a prototype edge system with a refined AIGC model deployment to implement and evaluate our LAD-TS method. DEdgeAI provides a real AIGC service for users, demonstrating up to 29.18% shorter service delays than the current five representative AIGC platforms. We release our open-source code at https://github.com/ChangfuXu/DEdgeAI/.

cs.LG↗

arXiv2024

Latent Neural Operator for Solving Forward and Inverse PDE Problems

Neural operators effectively solve PDE problems from data without knowing the explicit equations, which learn the map from the input sequences of observed samples to the predicted values. Most existing works build the model in the original geometric space, leading to high computational costs when the number of sample points is large. We present the Latent Neural Operator (LNO) solving PDEs in the latent space. In particular, we first propose Physics-Cross-Attention (PhCA) transforming representation from the geometric space to the latent space, then learn the operator in the latent space, and finally recover the real-world geometric space via the inverse PhCA map. Our model retains flexibility that can decode values in any position not limited to locations defined in the training set, and therefore can naturally perform interpolation and extrapolation tasks particularly useful for inverse problems. Moreover, the proposed LNO improves both prediction accuracy and computational efficiency. Experiments show that LNO reduces the GPU memory by 50%, speeds up training 1.8 times, and reaches state-of-the-art accuracy on four out of six benchmarks for forward problems and a benchmark for inverse problem. Code is available at https://github.com/L-I-M-I-T/LatentNeuralOperator.

cs.LG↗

arXiv2024

An effective open image theorem for products of principally polarized abelian varieties

Let $A = \prod_{1\leq i\leq n} A_i$ be the product of principally polarized abelian varieties $A_1, \ldots, A_n$ of dimensions $g_1, \ldots, g_n$, respectively, each defined over a number field $K$, and pairwise nonisogenous over $\overline{K}$. We make effective an open image theorem for $A$ due to Hindry and Ratazzi. More specifically, we give an explicit bound of the constant $c(A)$ under GRH, in terms of standard invariants of $K$ and each $A_i$, where $c(A)$ is defined to be the smallest positive integer such that for any prime $\ell>c(A)$, the image of the $\ell$-adic Galois representation of $A$ is "as large as possible" in a suitable sense.

math.NT↗

arXiv2024

Latent Neural Operator Pretraining for Solving Time-Dependent PDEs

Pretraining methods gain increasing attraction recently for solving PDEs with neural operators. It alleviates the data scarcity problem encountered by neural operator learning when solving single PDE via training on large-scale datasets consisting of various PDEs and utilizing shared patterns among different PDEs to improve the solution precision. In this work, we propose the Latent Neural Operator Pretraining (LNOP) framework based on the Latent Neural Operator (LNO) backbone. We achieve universal transformation through pretraining on hybrid time-dependent PDE dataset to extract representations of different physical systems and solve various time-dependent PDEs in the latent space through finetuning on single PDE dataset. Our proposed LNOP framework reduces the solution error by 31.7% on four problems and can be further improved to 57.1% after finetuning. On out-of-distribution dataset, our LNOP model achieves roughly 50% lower error and 3$\times$ data efficiency on average across different dataset sizes. These results show that our method is more competitive in terms of solution precision, transfer capability and data efficiency compared to non-pretrained neural operators.

cs.LG↗

arXiv2024

Opposing Average Congruence Class Biases in the Cyclicity and Koblitz Conjectures for Elliptic Curves

The cyclicity and Koblitz conjectures ask about the distribution of primes of cyclic and prime-order reduction, respectively, for elliptic curves over $\mathbb{Q}$. In 1976, Serre gave a conditional proof of the cyclicity conjecture, but the Koblitz conjecture (refined by Zywina in 2011) remains open. The conjectures are now known unconditionally "on average" due to work of Banks--Shparlinski and Balog--Cojocaru--David. Recently, there has been a growing interest in the cyclicity conjecture for primes in arithmetic progressions (AP), with relevant work by Akbal--Güloğlu and Wong. In this paper, we adapt Zywina's method to formulate the Koblitz conjecture for primes in AP and refine a theorem of Jones to establish results on the moments of the constants in both the cyclicity and Koblitz conjectures for AP. In doing so, we uncover a somewhat counterintuitive phenomenon: On average, these two constants are oppositely biased over congruence classes. Finally, in an accompanying repository, we give Magma code for computing the constants discussed in this paper.

math.NT↗

arXiv2024

Fed-Credit: Robust Federated Learning with Credibility Management

Aiming at privacy preservation, Federated Learning (FL) is an emerging machine learning approach enabling model training on decentralized devices or data sources. The learning mechanism of FL relies on aggregating parameter updates from individual clients. However, this process may pose a potential security risk due to the presence of malicious devices. Existing solutions are either costly due to the use of compute-intensive technology, or restrictive for reasons of strong assumptions such as the prior knowledge of the number of attackers and how they attack. Few methods consider both privacy constraints and uncertain attack scenarios. In this paper, we propose a robust FL approach based on the credibility management scheme, called Fed-Credit. Unlike previous studies, our approach does not require prior knowledge of the nodes and the data distribution. It maintains and employs a credibility set, which weighs the historical clients' contributions based on the similarity between the local models and global model, to adjust the global model update. The subtlety of Fed-Credit is that the time decay and attitudinal value factor are incorporated into the dynamic adjustment of the reputation weights and it boasts a computational complexity of O(n) (n is the number of the clients). We conducted extensive experiments on the MNIST and CIFAR-10 datasets under 5 types of attacks. The results exhibit superior accuracy and resilience against adversarial attacks, all while maintaining comparatively low computational complexity. Among these, on the Non-IID CIFAR-10 dataset, our algorithm exhibited performance enhancements of 19.5% and 14.5%, respectively, in comparison to the state-of-the-art algorithm when dealing with two types of data poisoning attacks.

cs.LG↗

arXiv2024

Quantitative upper bounds related to an isogeny criterion for elliptic curves

For $E_1$ and $E_2$ elliptic curves defined over a number field $K$, without complex multiplication, we consider the function ${\mathcal{F}}_{E_1, E_2}(x)$ counting non-zero prime ideals $\mathfrak{p}$ of the ring of integers of $K$, of good reduction for $E_1$ and $E_2$, of norm at most $x$, and for which the Frobenius fields $\mathbb{Q}(π_{\mathfrak{p}}(E_1))$ and $\mathbb{Q}(π_{\mathfrak{p}}(E_2))$ are equal. Motivated by an isogeny criterion of Kulkarni, Patankar, and Rajan, which states that $E_1$ and $E_2$ are not potentially isogenous if and only if ${\mathcal{F}}_{E_1, E_2}(x) = \operatorname{o} \left(\frac{x}{\log x}\right)$, we investigate the growth in $x$ of ${\mathcal{F}}_{E_1, E_2}(x)$. We prove that if $E_1$ and $E_2$ are not potentially isogenous, then there exist positive constants $κ(E_1, E_2, K)$, $κ'(E_1, E_2, K)$, and $κ''(E_1, E_2, K)$ such that the following bounds hold: (i) ${\mathcal{F}}_{E_1, E_2}(x) < κ(E_1, E_2, K) \frac{ x (\log\log x)^{\frac{1}{9}}}{ (\log x)^{\frac{19}{18}}}$; (ii) ${\mathcal{F}}_{E_1, E_2}(x) < κ'(E_1, E_2, K) \frac{ x^{\frac{6}{7}}}{ (\log x)^{\frac{5}{7}}}$ under the Generalized Riemann Hypothesis for Dedekind zeta functions (GRH); (iii) ${\mathcal{F}}_{E_1, E_2}(x) < κ''(E_1, E_2, K) x^{\frac{2}{3}} (\log x)^{\frac{1}{3}}$ under GRH, Artin's Holomorphy Conjecture for the Artin $L$-functions of number field extensions, and a Pair Correlation Conjecture for the zeros of the Artin $L$-functions of number field extensions.

math.NT↗

arXiv2024

Effective Gradient Sample Size via Variation Estimation for Accelerating Sharpness aware Minimization

Sharpness-aware Minimization (SAM) has been proposed recently to improve model generalization ability. However, SAM calculates the gradient twice in each optimization step, thereby doubling the computation costs compared to stochastic gradient descent (SGD). In this paper, we propose a simple yet efficient sampling method to significantly accelerate SAM. Concretely, we discover that the gradient of SAM is a combination of the gradient of SGD and the Projection of the Second-order gradient matrix onto the First-order gradient (PSF). PSF exhibits a gradually increasing frequency of change during the training process. To leverage this observation, we propose an adaptive sampling method based on the variation of PSF, and we reuse the sampled PSF for non-sampling iterations. Extensive empirical results illustrate that the proposed method achieved state-of-the-art accuracies comparable to SAM on diverse network architectures.

cs.CV↗