Haijun Zhang

Yanjie Dong, Luya Wang, Yuanfang Chi et al.

A wireless federated learning system is investigated by allowing a server and workers to exchange uncoded information via orthogonal wireless channels. Since the workers frequently upload local gradients to the server via bandwidth-limited channels, the uplink transmission from the workers to the server becomes a communication bottleneck. Therefore, a one-shot distributed principle component analysis (PCA) is leveraged to reduce the dimension of uploaded gradients such that the communication bottleneck is relieved. A PCA-based wireless federated learning (PCA-WFL) algorithm and its accelerated version (i.e., PCA-AWFL) are proposed based on the low-dimensional gradients and the Nesterov's momentum. For the non-convex loss functions, a finite-time analysis is performed to quantify the impacts of system hyper-parameters on the convergence of the PCA-WFL and PCA-AWFL algorithms. The PCA-AWFL algorithm is theoretically certified to converge faster than the PCA-WFL algorithm. Besides, the convergence rates of PCA-WFL and PCA-AWFL algorithms quantitatively reveal the linear speedup with respect to the number of workers over the vanilla gradient descent algorithm. Numerical results are used to demonstrate the improved convergence rates of the proposed PCA-WFL and PCA-AWFL algorithms over the benchmarks.

Yanhang Shi, Siguang Chen, Haijun Zhang

Since federated learning (FL) has been introduced as a decentralized learning technique with privacy preservation, statistical heterogeneity of distributed data stays the main obstacle to achieve robust performance and stable convergence in FL applications. Model personalization methods have been studied to overcome this problem. However, existing approaches are mainly under the prerequisite of fully labeled data, which is unrealistic in practice due to the requirement of expertise. The primary issue caused by partial-labeled condition is that, clients with deficient labeled data can suffer from unfair performance gain because they lack adequate insights of local distribution to customize the global model. To tackle this problem, 1) we propose a novel personalized semi-supervised learning paradigm which allows partial-labeled or unlabeled clients to seek labeling assistance from data-related clients (helper agents), thus to enhance their perception of local data; 2) based on this paradigm, we design an uncertainty-based data-relation metric to ensure that selected helpers can provide trustworthy pseudo labels instead of misleading the local training; 3) to mitigate the network overload introduced by helper searching, we further develop a helper selection protocol to achieve efficient communication with acceptable performance sacrifice. Experiments show that our proposed method can obtain superior performance and more stable convergence than other related works with partially labeled data, especially in highly heterogeneous setting.

Haojin Li, Anbang Zhang, Chen Sun et al.

The low-altitude economy (LAE) is rapidly expanding driven by urban air mobility, logistics drones, and aerial sensing, while fast and accurate beam prediction in uncrewed aerial vehicles (UAVs) communications is crucial for achieving reliable connectivity. Current research is shifting from single-signal to multi-modal collaborative approaches. However, existing multi-modal methods mostly employ fixed or empirical weights, assuming equal reliability across modalities at any given moment. Indeed, the importance of different modalities fluctuates dramatically with UAV motion scenarios, and static weighting amplifies the negative impact of degraded modalities. Furthermore, modal mismatch and weak alignment further undermine cross-scenario generalization. To this end, we propose a reliability-aware dynamic weighting scheme applied to a semantic-aware multi-modal beam prediction framework, named SaM2B. Specifically, SaM2B leverages lightweight cues such as environmental visual, flight posture, and geospatial data to adaptively allocate contributions across modalities at different time points through reliability-aware dynamic weight updates. Moreover, by utilizing cross-modal contrastive learning, we align the "multi-source representation beam semantics" associated with specific beam information to a shared semantic space, thereby enhancing discriminative power and robustness under modal noise and distribution shifts. Experiments on real-world low-altitude UAV datasets show that SaM2B achieves more satisfactory results than baseline methods.

Ning Yang, Shuo Chen, Haijun Zhang et al.

Mobile Edge Computing (MEC) broadens the scope of computation and storage beyond the central network, incorporating edge nodes close to end devices. This expansion facilitates the implementation of large-scale "connected things" within edge networks. The advent of applications necessitating real-time, high-quality service presents several challenges, such as low latency, high data rate, reliability, efficiency, and security, all of which demand resolution. The incorporation of reinforcement learning (RL) methodologies within MEC networks promotes a deeper understanding of mobile user behaviors and network dynamics, thereby optimizing resource use in computing and communication processes. This paper offers an exhaustive survey of RL applications in MEC networks, initially presenting an overview of RL from its fundamental principles to the latest advanced frameworks. Furthermore, it outlines various RL strategies employed in offloading, caching, and communication within MEC networks. Finally, it explores open issues linked with software and hardware platforms, representation, RL robustness, safe RL, large-scale scheduling, generalization, security, and privacy. The paper proposes specific RL techniques to mitigate these issues and provides insights into their practical applications.

Wuxuan Jiang, Xiangjun Song, Shenbai Hong et al.

Accuracy and efficiency remain challenges for multi-party computation (MPC) frameworks. Spin is a GPU-accelerated MPC framework that supports multiple computation parties and a dishonest majority adversarial setup. We propose optimized protocols for non-linear functions that are critical for machine learning, as well as several novel optimizations specific to attention that is the fundamental unit of Transformer models, allowing Spin to perform non-trivial CNNs training and Transformer inference without sacrificing security. At the backend level, Spin leverages GPU, CPU, and RDMA-enabled smart network cards for acceleration. Comprehensive evaluations demonstrate that Spin can be up to $2\times$ faster than the state-of-the-art for deep neural network training. For inference on a Transformer model with 18.9 million parameters, our attention-specific optimizations enable Spin to achieve better efficiency, less communication, and better accuracy.

Peijun Zhu, Ning Yang, Jiayu Wei et al.

Mixture-of-Experts (MoE) Large Language Models (LLMs) face a trilemma of load imbalance, parameter redundancy, and communication overhead. We introduce a unified framework based on dynamic expert clustering and structured compression to address these issues cohesively. Our method employs an online clustering procedure that periodically regroups experts using a fused metric of parameter and activation similarity, which stabilizes expert utilization. To our knowledge, this is one of the first frameworks to leverage the semantic embedding capability of the router to dynamically reconfigure the model's architecture during training for substantial efficiency gains. Within each cluster, we decompose expert weights into a shared base matrix and extremely low-rank residual adapters, achieving up to fivefold parameter reduction per group while preserving specialization. This structure enables a two-stage hierarchical routing strategy: tokens are first assigned to a cluster, then to specific experts within it, drastically reducing the routing search space and the volume of all-to-all communication. Furthermore, a heterogeneous precision scheme, which stores shared bases in FP16 and residual factors in INT4, coupled with dynamic offloading of inactive clusters, reduces peak memory consumption to levels comparable to dense models. Evaluated on GLUE and WikiText-103, our framework matches the quality of standard MoE models while reducing total parameters by approximately 80%, improving throughput by 10% to 20%, and lowering expert load variance by a factor of over three. Our work demonstrates that structural reorganization is a principled path toward scalable, efficient, and memory-effective MoE LLMs.

Tuo Zhang, Ning Li, Xin Yuan et al.

With the breakthrough progress of large language models (LLMs) in natural language processing and multimodal tasks, efficiently deploying them on resource-constrained edge devices has become a critical challenge. The Mixture of Experts (MoE) architecture enhances model capacity through sparse activation, but faces two major difficulties in practical deployment: (1) The presence of numerous outliers in activation distributions leads to severe degradation in quantization accuracy for both activations and weights, significantly impairing inference performance; (2) Under limited memory, efficient offloading and collaborative inference of expert modules struggle to balance latency and throughput. To address these issues, this paper proposes an efficient MoE edge deployment scheme based on Hessian-Aware Quantization (HAQ) and CPU-GPU collaborative inference. First, by introducing smoothed Hessian matrix quantization, we achieve joint 8-bit quantization of activations and weights, which significantly alleviates the accuracy loss caused by outliers while ensuring efficient implementation on mainstream hardware. Second, we design an expert-level collaborative offloading and inference mechanism, which, combined with expert activation path statistics, enables efficient deployment and scheduling of expert modules between CPU and GPU, greatly reducing memory footprint and inference latency. Extensive experiments validate the effectiveness of our method on mainstream large models such as the OPT series and Mixtral 8*7B: on datasets like Wikitext2 and C4, the inference accuracy of the low-bit quantized model approaches that of the full-precision model, while GPU memory usage is reduced by about 60%, and inference latency is significantly improved.