Large language model (LLM) services increasingly rely on distributed inference across multiple GPU servers to sustain concurrent requests under limited compute, memory, and bandwidth resources. In such settings, a partitioned LLM can be represented as an inference chain (InFC), where the deployment decision determines both the sustainable concurrency ceiling (SCC) on the revenue side and the memory and communication overhead on the cost side. This paper studies the profit-aware inference chain deployment (InFCD) problem in heterogeneous data center networks. We show that increasing the InFC length does not monotonically improve profit: finer partitioning can relieve per-GPU resource bottlenecks and improve SCC, but may also increase deployment spread, inference path length, and internal traffic. To capture this tradeoff, we formulate profit-aware InFCD by jointly modeling static-weight vRAM occupation, per-user KV-cache occupation, user-side traffic, internal boundary traffic, and resource-coupled SCC, and prove its NP-hardness. We then propose the Maximum Sub-module Deployment Gain (MSDG) score and design an MSDG-based greedy algorithm. Theoretical analysis characterizes its online complexity and establishes a conditional positive-profit preservation property. Simulations show that MSDG improves total profit over SCC-oriented, cost-oriented, and local-profit-oriented baselines, characterize empirical optimality gaps and SLO sensitivity.

The accuracy of digital twin models hinges on the prompt collection of information from the vehicular environment. However, the high mobility of vehicles and the dynamically changing network environment pose significant challenges. Dynamic twin model migration can reduce the Age of Information (AoI) by bringing twin models closer to their vehicles. Existing works rarely consider the inherent differences in optimization cycles between digital twin model migration and data upload, which potentially leads to suboptimal cost efficiency and information freshness. Specifically, real-time vehicular data must be rapidly uploaded to edge servers to ensure the accuracy and timeliness of digital twin models, while frequent migration of twin models over short periods incurs substantial costs. Therefore, we propose a dual-timescale bilevel learning approach, where the upper-layer learning optimizes twin model migration decisions on a long timescale to achieve forward-looking model migration, and the lower-layer learning optimizes data upload and resource allocation decisions on a short timescale to ensure the accuracy and timeliness of digital twin models. Then, we design a multi-agent selective parameter sharing approach based on spatiotemporal dependency correlations to accelerate model convergence and reduce communication costs among agents. Furthermore, through a rigorous theoretical analysis, we prove the convergence of the dual-timescale bilevel learning with broad applicability. Finally, numerical results demonstrate that our algorithm outperforms comparison algorithms in terms of convergence, AoI, and system cost, achieving at least a 21.30% reduction in AoI and a 14.58% reduction in system cost compared to the benchmark algorithms.

Mobile Edge Computing (MEC) decentralizes the network's computing and storage capabilities from centralized infrastructure to edge nodes located closer to end-users, enabling context-aware service deployment, low-latency service response, and efficient computation for mobile users. However, achieving energy-efficient service deployment while maintaining service delivery quality remains a significant challenge due to the wide geographic distribution of edge nodes, the dynamic variation of service workloads, and the differences between service deployment and delivery cycles. To address these challenges, we first design a feature encoding strategy and a self-attention-based encoder to extract contextual features, which are fused to support adaptive decision timescale regulation driven by service semantics and system load dynamics. Then, we propose a novel Dual-Timescale Energy-Efficient Service Deployment and Delivery (DT-EESD) framework integrated with hierarchical learning. The upper layer leverages an enhanced decision-making mechanism to optimize proactive service deployment and base station switching on a larger timescale, aiming to reduce long-term network costs. The lower layer employs a fine-grained real-time optimization approach to dynamically handle service delivery and resource allocation on a smaller timescale, effectively responding to dynamic service requests. By incorporating an expected reward-based learning mechanism, the framework efficiently handles the temporal coupling between deployment and delivery cycles. Extensive experiments demonstrate that DT-EESD outperforms baseline algorithms, achieving at least a 10.89% reduction in average system cost while also improving model convergence, reducing delay, and enhancing resource utilization.

Deploying ultra-reliable and low-latency communication (URLLC) service function chains (SFCs) is imperative for applications demanding stringent latency and reliability performances. In these applications, ensuring uninterrupted service hinges on establishing fault-disjoint primary and backup service function paths (SFPs). However, existing techniques for deploying SFCs fall short in optimizing latency differentials between the primary and backup SFPs, posing risks of service disruptions in critical URLLC applications like remote surgery, smart factory, and unmanned vehicle systems. In this work, we investigate pioneering techniques to efficiently optimize the primary and backup SFP latencies while minimizing their differentials. We formally formulate the problem of ultra-reliable and low-latency SFC deployment (URLLC-SD) and show its NPhardness. We develop an innovative algorithm, the Yen-based SFP Identification in Layered Graph (YANG), which optimizes the equal-weight composite latency objectives with symmetric QoS/SLA for primary and backup SFPs at the expense of runtime complexity. Through extensive simulations, we demonstrate the YANG's superiority, surpassing state-of-the-art benchmarks. In particular, YANG achieves the highest acceptance rates under specific constraints on SFP latency and latency differentials. Furthermore, our analysis reveals interesting insights into the selection of good path candidates to optimize URLLC-SD.

Numerous existing deep learning models for time-series classification (TSC) tend to overlook the intricate interplay between higher-and lower-level semantic information. While the focus is often on extracting higher-level semantics from lower-level sources, the reciprocal influence of lower-level information on higher levels is undervalued. To address this, we propose an ensemble transitive bidirectional decoupled self-distillation (ETBiDecSD) method for TSC. ETBiDecSD enhances the robustness of higher-level semantic information using an average feature ensemble (AFE) method to amalgamate the output from each level. Simultaneously, the integrated features are transmitted to each lower level through a directional decoupled distillation (DD) structure. Additionally, to promote deep interaction between higher-and lower-level semantic information, ETBiDecSD introduces a transitive bidirectional DD (TBDD) structure, facilitating the transfer of target-class and nontarget-class knowledge between higher and lower levels. Experimental results demonstrate that whether a fully convolutional network (FCN) with four convolutional blocks or InceptionTime with four Inception blocks is used as the baseline, ETBiDecSD outperforms a quantity of well-established self-distillation algorithms across 85 widely used UCR2018 datasets, as evidenced by the metrics “win”/“tie”/“lose” and avg. rank, which are derived from accuracy andF1-scores. Notably, when compared to a nonself-distillation FCN, ETBiDecSD achieves “win”/“tie”/“lose” results of 64/4/17 in terms of accuracy and 65/4/16 in terms ofF1-score. Similarly, in comparison to a nonself-distillation InceptionTime, ETBiDecSD attains “win”/“tie”/“lose” results of 60/12/13 for accuracy and 57/12/16 forF1-score.

Edge-based video analytics faces the dual challenge of dynamic network conditions and strict real-time processing requirements. Existing offloading and scheduling methods often fail to adapt efficiently to varying link capacities, resulting in high latency, unstable detection accuracy, and frame loss. In this paper, we propose LDKite, a reinforcement learning-driven, linkadaptive framework for real-time video analytics in dynamic edge networks. LDKite jointly optimizes computation placement and frame offloading decisions by continuously learning from network state feedback, enabling adaptive scheduling that accounts for both network bandwidth variations and task-specific processing demands. The framework incorporates a lightweight policy network that efficiently guides offloading decisions while maintaining scalability to large edge networks. Extensive experiments on realistic edge scenarios demonstrate that LDKite outperforms state-of-the-art baselines, achieving up to 25% reduction in frame processing delay, only 4% frame loss, and 91% detection accuracy. The results validate LDKite's effectiveness in delivering robust, scalable, and high-performance video analytics under dynamic edge conditions, highlighting its potential for deployment in real-world edge computing environments.

With the advancement of Intelligent Transportation Systems (ITS), Digital Twin (DT) technology has been widely applied to tasks such as traffic flow modeling and autonomous driving assistance. However, traditional standalone DT models often suffer from weak generalization ability and potential risks of privacy leakage in data-drifting scenarios. To address these challenges, a novel FL-DTN architecture for vehicular networks is proposed by integrating Federated Learning (FL) with Digital Twin Networking (DTN), aiming to preserve data privacy and enhance generalization ability of digital twin models. Specifically, an online resource allocation algorithm, Online Resource Allocation for Synchronization, Retraining and Inference (ORASRI), is designed to dynamically balance the resource allocation among digital twin synchronization, retraining, and inference for Vehicle Digital Twins (VDTs), while adapting to data drift under constrained resource conditions. In addition, a Teacher-Student collaborative mechanism is introduced to improve inference accuracy while reducing resource consumption. Experiments on the MNIST-C dataset show that ORASRI improve 4.6% and 10.3% Inference accuracy in non-FL and FL data-drifting scenarios, respectively.

Synchronization of vehicular digital twin (VDT) state data is essential for maintaining the accuracy of digital twin models. Existing studies show that VDT data synchronization typically requires a substantial amount of bandwidth and frequent data exchanges. However, real-world vehicle state data are prone to noise interference and bandwidth constraints, causing synchronization errors and significantly degrading VDT accuracy. To evaluate the impact of noise interference on VDT accuracy, we model vehicle state evolution as a discrete-time wiener process and employ a Kalman filter for optimal state estimation. By further analyzing the relationship between estimation error and update timeliness, we find that weighted scheduling based on Age of Information (AoI) effectively suppresses error accumulation and improves synchronization performance. Then, we propose an aoi error-aware scheduling mechanism maximum weighted noise age (MWNA), within a cloud-edge collaborative VDT framework, MWNA dynamically evaluates each vehicle’s state update and prioritizes the transmissions that yield the greatest reduction in estimation error. Compared with baselines, the MWNA policy achieves up to 25–40% lower average error across a range of correlation settings, particularly under high noise correlation and large-scale vehicular scenarios.

Cloud-based large language models (LLMs) have gained widespread adoption among human users. However, when the users shift to Internet-enabled machines, centralized LLM systems often suffer from high latency and fail to provide timely responses. Deploying LLMs over edge networks presents a promising alternative, yet maintaining an up-to-date knowledge base to address the dynamic and time-sensitive demands of diverse machine users remains a significant challenge. Consequently, there is a pressing need for fast and cost-efficient LLM deployment strategies tailored to edge environments. In this work, we address the problem of deploying mixture-of-experts (MoE) LLMs in a runtime- and cost-efficient manner. We formally define the Expert Model Deployment in Edge Networks (EMD-EN) problem, aiming to minimize deployment costs. Leveraging the inherent modularity of MoE, we propose a novel Neighbor-First Centrality (NFC) metric to facilitate the placement of model components across edge nodes and design the NFC-based Mixture of Expert layer Deployment (NFC-MoED) algorithm. Our results show that NFC-MoED substantially improves runtime efficiency and maintains near-optimal deployment costs compared to the brute-force benchmark.

To support edge intelligence, knowledge distillation (KD) is widely employed to compress large language models (LLMs) into smaller, domain-specific student models. However, due to the limited generalization capabilities of student models, they may fail to provide accurate responses across diverse domains. In such cases, the teacher model serves as a complementary component, handling queries that exceed the scope of student models. In line with the KD paradigm, this work investigates a collaborative deployment framework in which multiple student models are distributed across the network edge to serve the majority of client requests. In contrast, a centralized teacher model addresses more complex or ambiguous queries. To begin, we formally define the Student Model Placement in Edge Networks (SMP-EN) problem, aiming to minimize total access costs. We prove that SMP-EN is NP-hard, and to address this challenge, we introduce an Access Cost Measure (ACM) that quantifies the expected access costs. Building upon this measure, we propose the ACM-based Student Model Placement (ACM-SMP) algorithm to determine student model placement efficiently. Extensive simulations show that ACM-SMP significantly reduces the average expected client access cost compared to benchmarks.

With its rapid development, AIGC applications have expanded to diverse generative content, including text, images, audio, and videos. To enhance the AIGC's output quality, prompt chains are proposed to structure the generation process. Owing to the prompt's data processing nature, one compromised prompt engineering-enabled server (PES) may propagate vulnerabilities across networks, leading to unintended content generation, system risks, and potential user trust issues. To mitigate these risks, this work adopts a zero-trust (ZT) security framework to protect inter-server communications. We define the problem of prompt chain deployment in ZT-enabled networks(PCD-ZT), with the objective of minimizing total service costs, encompassing both deployment and ZT verification costs. To address this, we introduce the verification-cost-balance (VCB) factor that helps reduce ZT verifications to save the overall service cost and accordingly propose an algorithm called sub-prompt-chain brand and bound (SCBB). Extensive simulations demonstrate that our SCBB achieves overall service cost reductions of 4.67 % and 13.89 %, and cuts verification costs by 11.86 % and 30.50 %, compared to the benchmarks extended from the state-of-the-art.

Recent advances in large language models (LLMs) have empowered robots and drones with autonomous decision-making capabilities. Due to the stringent real-time requirements of these applications, LLM inference must be performed at the network edge. However, hosting high-precision LLMs on a single edge server is often infeasible, creating challenges in efficiently distributing LLM deployments across edge networks. This work addresses these challenges by formulating and solving the profit maximization problem for distributed LLM inference deployment. We first formally define the Profit-Centric Inference Chain Deployment (PC-InCD) problem. To solve PC-InCD, we introduce a novel Local Maximal Profit (LMP) factor that enables effective edge server selection for hosting LLM sub-modules, and we propose the LMP-based Inference Chain Deployment (LMP-InCD) algorithm. Extensive simulations demonstrate that LMP-InCD significantly outperforms benchmark methods in maximizing profit across diverse network conditions.

BACKGROUND: Drug response prediction is critical in precision medicine to determine the most effective and safe treatments for individual patients. Traditional prediction methods relying on demographic and genetic data often fall short in accuracy and robustness. Recent graph-based models, while promising, frequently neglect the critical role of atomic interactions and fail to integrate drug fingerprints with SMILES for comprehensive molecular graph construction. RESULTS: We introduce multimodal multi-channel graph attention network with adaptive fusion (MGATAF), a framework designed to enhance drug response predictions by capturing both local and global interactions among graph nodes. MGATAF improves drug representation by integrating SMILES and fingerprints, resulting in more precise predictions of drug effects. The methodology involves constructing multimodal molecular graphs, employing multi-channel graph attention networks to capture diverse interactions, and using adaptive fusion to integrate these interactions at multiple abstraction levels. Empirical results demonstrate MGATAF's superior performance compared to traditional and other graph-based techniques. For example, on the GDSC dataset, MGATAF achieved a 5.12% improvement in the Pearson correlation coefficient (PCC), reaching 0.9312 with an RMSE of 0.0225. Similarly, in new cell-line tests, MGATAF outperformed baselines with a PCC of 0.8536 and an RMSE of 0.0321 on the GDSC dataset, and a PCC of 0.7364 with an RMSE of 0.0531 on the CCLE dataset. CONCLUSIONS: MGATAF significantly advances drug response prediction by effectively integrating multiple molecular data types and capturing complex interactions. This framework enhances prediction accuracy and offers a robust tool for personalized medicine, potentially leading to more effective and safer treatments for patients. Future research can expand on this work by exploring additional data modalities and refining the adaptive fusion mechanisms.

Multi-modal semantic communication (MSC) has gained great attention due to its multi-modal processing ability. However, the existing MSC systems are mainly built on multi-modal large models that lead to inefficient computation on non-essential tokens, potentially restricting MSC from achieving more advanced levels of intelligence. To address this challenge, we propose a large model-empowered MSC system with a cross-modal attention-based token selection mechanism, denoted as LMECM-SC, which effectively utilizes the attention score across multi-modal tokens to filter out noisy or unuseful tokens, selectively learning the tokens that best benefit downstream applications. Meanwhile, we introduce the multi-modal adaptive semantic encoder and decoder that dynamically assign weights to encode multi-modal semantic information extracted from the selected tokens based on their modality and integrate semantic information with cross-modal attention scores at the receiver, optimizing the performance on downstream tasks. Experiment results indicate that LMECM-SC effectively reduces the number of tokens used for training, outperforming four baseline methods in terms of bilingual evaluation understudy score for text, learned perceptual image patch similarity for image, and perceptual evaluation of speech quality score for speech.

Over the years, various sophisticated deep learning algorithms have surfaced for multivariate time series classification (MTSC), notably the dual-network-based model. This model comprises two parallel networks tailored to time series data: one for local feature extraction and the other for global relation extraction. However, effectively integrating these dual networks poses a significant challenge. To address this, we propose a knowledge aggregation transformer network (KATN) for MTSC. KATN, composed of four aggregation transformer blocks, extracts abundant regularizations and connections hidden within the data. Each block incorporates a modified residual network (MResNet) for local feature extraction and a multi-head attention network for global relation extraction. Initially, the block merges MResNet's output feature with that of the multi-head attention network through an additive operation. Subsequently, it aligns features with a fully connected (i.e., dense) layer and activates neural units using the Gaussian error linear unit function. This strategic feature aggregation allows for capturing long-range dependencies among multiple variables in multivariate time series data. Experimental results demonstrate that KATN significantly outperforms 6 state-of-the-art transformer variants, achieving a ‘win’/‘tie’/‘lose’ record of 9/6/15 and securing the lowest AVG_rank score. Furthermore, when evaluated against 18 existing MTSC algorithms across 13 UEA datasets, KATN consistently delivers superior performance, attaining the lowest AVG_rank score among all compared methods.

The unlawful use of noncooperative drones, or unmanned aerial vehicles (UAVs), poses serious threats to public safety and societal security, necessitating robust monitoring solutions. However, the detection of drones, particularly those flying remotely, is often hindered by the limited accuracy in identifying small targets. Additionally, challenges related to insufficient lightweight design and complexities in practical deployment further exacerbate the issue. To address these challenges, we propose a quad-head network, QHNet, designed to provide scalable, adaptable, and efficient drone detection across diverse scenarios. QHNet is available in five scalable model sizes—Nano (N), Small (S), Medium (M), Large (L), and Extra Large (X)—ensuring real-time detection tailored to varying resource constraints and operational requirements. The core of QHNet lies in its innovative quad detection head (QDH), which introduces an additional detection layer to perform secondary feature extraction for small targets, enhancing detection precision. Furthermore, the architecture integrates adaptive spatial feature fusion to improve scale invariance and overall detection accuracy. To further optimize performance, QHNet incorporates a four-scale feature fusion network (FSF) and a specialized small target IoU loss function (STIoU) to refine detection precision. Simultaneously, the lightweight coarse-to-fine processing unit (LCF) and SCDown downsampling (SCD) form a comprehensive lightweight scheme, significantly reducing computational overhead while maintaining high detection efficacy. Extensive experiments were conducted on DUT-Plus, an augmented version of the DUT Anti-UAV dataset enhanced through data augmentation. The experimental results confirm that QHNet attains outstanding performance, providing an advanced balance between detection accuracy and computational efficiency across all model sizes.

This paper studies the joint optimization of multi-objective dependent task scheduling, resource allocation, and service caching in an aerial-ground integrated mobile edge computing system that includes multiple uncrewed aerial vehicles (UAVs). These UAVs, in coordination with a high-altitude platform, work together to process numerous dependent tasks collected by the UAVs. The optimization problem involves two conflicting objectives that need to be minimized simultaneously: the average execution delay of all dependent tasks and the average energy consumption of all UAVs. The conflict between the two objectives makes the problem quite challenging. Recently, some multi-objective approaches, such as multi-objective evolutionary algorithms (MOEAs), have been introduced to address dependent task scheduling. However, these approaches often suffer from premature convergence and tend to fall into local optima. To address these issues, we propose a modified MOEA based on decomposition that incorporates two performance-improving strategies. The first one is a probability-based neighborhood search strategy that selects two individuals to update neighborhood individuals based on the neighborhoods and external population, thereby improving population updating efficiency. The second one is a dynamic voltage and frequency scaling-based energy reduction strategy that further enhances the quality of solutions by adjusting the computing frequencies. Experimental results verify that the proposed algorithm obtains a number of outstanding nondominated solutions and achieves a better balance between objectives compared with several algorithms.

EdgeAI represents a compelling approach for deploying DNN models at network edge through model partitioning. However, most existing partitioning strategies have primarily concentrated on homogeneous environments, neglecting the effect of device placement and their inapplicability to heterogeneous settings. Moreover, these strategies often rely on either data parallelism or model parallelism, each presenting its own limitations, including data synchronization and communication overhead. This paper aims at enhancing inference performance through a pipeline system of devices through leveraging both parallel and sequential relationships among them. Accordingly, the problem of Multi-Device Cooperative DNN Inference is formulated by optimizing both device placement and model partitioning, taking into account the unique characteristics of heterogeneous edge resources and DNN models, with the goal of maximizing throughput. To this end, we propose an evolutionary device placement technique to determine the pipeline stage of devices by enhancing a variant of particle swarm optimization. Subsequently, an adaptive model partitioning strategy is developed by combining intra-layer and inter-layer model partitioning based on dynamic programming and the input-output mapping of DNN layers, respectively, to accommodate edge resource limitations. Finally, we construct a simulation model and a prototype, and the extensive results demonstrate that our proposed algorithm outperforms current state-of-the-art algorithms.

Nowadays, cross-device federated learning (FL) is the key to achieving personalization services for mobile users and has been widely employed by companies like Google, Microsoft, and Alibaba in production. With the explosive growth in the number of participants, the central FL server, which acts as the manager and aggregator of cross-device model training, would get overloaded, becoming the system bottlenecks. Inspired by the emerging wave of edge computing, an interesting question arises:Could edge clouds help cross-device FL systems overcome the bottleneck?This article provides a cautiously optimistic answer by proposingINP, a FL-specific In-Network Processing framework to achieve the goal. As in-network processing has broken the end-to-end principle of the involved communication and lacks the support of transport protocols, the key is to design domain-specific transport protocols forINP. To fill the gap, we propose the novel Model Download Protocol ofmdpand Model Upload Protocol ofmup. Withmdpandmup, edge cloud nodes along the paths inINPcan easily eliminate duplicated model downloads and pre-aggregate associated gradient uploads for the central FL server, thus alleviating its bottleneck effect, and further accelerating the entire training progress significantly.

Semantic communication (SC) has emerged as a promising paradigm for upcoming intelligent applications, enabling mobile devices to collaboratively execute intelligent tasks with edge servers through computation offloading. However, few studies have addressed the problem of collaborative inference in SC networks. Traditional collaborative inference mechanisms may suffer performance decline in SC systems and are vulnerable to eavesdroppers. To address these issues, first, we present an encryptor that encrypts semantic information to avoid privacy leakage and a decryptor for restoration. Besides, we propose a novel SC-empowered edge computing framework enabling mobile devices to deploy a partial semantic encoder and offload the rest to edge servers. Based on this framework, we formulate the collaborative inference optimization problem, jointly optimizing delay, energy consumption, and privacy leakage. DNNPart is devised based on deep deterministic policy gradient to address the problem, which consists of a semantic attention mechanism that enables it to focus on important state variables, a hybrid action representation method that makes it adapt to mixed discrete and continuous action spaces, a dynamic model splitting algorithm that locates the optimal partition layer and adaptively splits the semantic coders. Integrated with these components, DNNPart iteratively optimizes the offloading strategy to find the optimal offloading strategy. Extensive simulations were conducted to verify the effectiveness of the proposed method by comparing it with baseline mechanisms.

This article proposes a semi-supervised contrastive capsule transformer method with feature-based knowledge distillation (KD) that simplifies the existing semisupervised learning (SSL) techniques for wearable human activity recognition (HAR), called CapMatch. CapMatch gracefully hybridizes supervised learning and unsupervised learning to extract rich representations from input data. In unsupervised learning, CapMatch leverages the pseudolabeling, contrastive learning (CL), and feature-based KD techniques to construct similarity learning on lower and higher level semantic information extracted from two augmentation versions of the data, "weak" and "timecut," to recognize the relationships among the obtained features of classes in the unlabeled data. CapMatch combines the outputs of the weak- and timecut-augmented models to form pseudolabeling and thus CL. Meanwhile, CapMatch uses the feature-based KD to transfer knowledge from the intermediate layers of the weak-augmented model to those of the timecut-augmented model. To effectively capture both local and global patterns of HAR data, we design a capsule transformer network consisting of four capsule-based transformer blocks and one routing layer. Experimental results show that compared with a number of state-of-the-art semi-supervised and supervised algorithms, the proposed CapMatch achieves decent performance on three commonly used HAR datasets, namely, HAPT, WISDM, and UCI_HAR. With only 10% of data labeled, CapMatch achieves values of higher than 85.00% on these datasets, outperforming 14 semi-supervised algorithms. When the proportion of labeled data reaches 30%, CapMatch obtains values of no lower than 88.00% on the datasets above, which is better than several classical supervised algorithms, e.g., decision tree and -nearest neighbor (KNN).

Recently, numerous deep learning algorithms have addressed wearable human activity recognition (HAR), but they often struggle with efficient knowledge transfer to lightweight models for mobile devices. Knowledge distillation (KD) is a popular technique for model compression, transferring knowledge from a complex teacher to a compact student. Most existing KD algorithms consider homogeneous architectures, hindering performance in heterogeneous setups. This is an under-explored area in wearable HAR. To bridge this gap, we propose a heterogeneous mutual KD (HMKD) framework for wearable HAR. HMKD establishes mutual learning within the intermediate and output layers of both teacher and student models. To accommodate substantial structural differences between teacher and student, we employ a weighted ensemble feature approach to merge the features from their intermediate layers, enhancing knowledge exchange within them. Experimental results on the HAPT, WISDM, and UCI_HAR datasets show HMKD outperforms ten state-of-the-art KD algorithms in terms of classification accuracy. Notably, with ResNetLSTMaN as the teacher and MLP as the student, HMKD increases by 9.19% in MLP's $F_{1}$ score on the HAPT dataset.

Recent advances in distributed machine learning show theoretically and empirically that, for many models, provided that workers will eventually participate in the synchronizations,$i)$the training still converges, even if only$p$workers take part in each round of synchronization, and$ii)$a larger$p$generally leads to a faster rate of convergence. These findings shed light on eliminating the bottleneck effects of parameter synchronization in large-scale data-parallel distributed training and have motivated several optimization designs. In this paper, we focus on optimizing the parameter synchronization forpeer-to-peerdistributed learning, where workers broadcast or multicast their updated parameters to others for synchronization, and proposeSelMcast, a suite of expressive and efficient multicast receiver selection algorithms, to achieve the goal. Compared with the state-of-the-art (SOTA) design, which randomly selects exactly$p$receivers for each worker’s multicast in a bandwidth-agnostic way,SelMcastchooses receivers based on the global view of their available bandwidth and loads, yielding two advantages, i.e., accelerated parameter synchronization for higher utilization of computing resources and enlarged average$p$values for faster convergence. Comprehensive evaluations show thatSelMcastis efficient for both peer-to-peer Bulk Synchronous Parallel (BSP) and Stale Synchronous Parallel (SSP) distributed training, outperforming the SOTA solution significantly.

This paper studies a distributed semantic communication (SC) problem for multivariate time series forecasting tasks in edge environments, with heterogeneous clients considered. At the client side, a semantic encoder is composed of a number of federated blocks and this number is subject to local resource availability. Each federated block consists of a patch-wise attention module (PAM) and a federated adapter, extracting semantic information for efficient transmission across wireless channels. Based on the federated adapters, this paper proposes an SC-oriented heterogeneous federated learning architecture, named SC-FedAda. SC-FedAda adopts self-distillation to facilitate cross-client and cross-layer knowledge sharing, enabling efficient collaborative inference. At the edge server, semantic signals are fed into a channel decoder and then a semantic decoder. The semantic decoder consists of a PAM and a fully connected network for forecasting tasks. Simulation results demonstrate that SC-FedAda outperforms four state-of-the-art federated learning-based structures under three types of wireless channels, i.e. SC-FedAda achieves much lower forecasting loss on three widely-used time series datasets, particularly in low signal-to-noise ratio scenarios.

Mobile edge computing (MEC) decentralizes the computational and storage capabilities of the network to edge nodes, providing support for the dynamic deployment and rapid response of mobile services. However, the large-scale distributed deployment of edge nodes, their widespread geographic distribution, multi-dimensional and complexly dependent service characteristics, and the dynamically changing network environment pose challenges to energy-efficient service deployment. In this paper, we consider multi-dimensional features for service deployment, including dynamic traffic demand, geography information, service semantics, and service popularity, with the aim of improving the availability of edge services and reducing network energy consumption. Firstly, to handle the large volume of edge service data, we design a multi-dimensional feature extraction approach based on the Transformer model, which does not rely on the sequential order of data and can enhance computational efficiency of edge models through parallel processing. Then, to adapt to the dynamically changing edge network environment, we propose an Energy-Efficient Service Deployment algorithm (EESD) based on the improved Dueling Deep Q-Network, which makes service deployment and base station switching decisions in a learning-based manner. Finally, simulation results demonstrate that EESD outperforms comparison algorithms in terms of model convergence, system total cost, and energy consumption.

Cooperative perception, based on vehicle-to-everything (V2X) communication technology, is a promising solution for connected and automated vehicles (CAVs) to improve their perception capabilities in intelligent transportation systems. The frequency of message transmission in cooperative perception among mobile vehicles plays a crucial role, as it directly impacts communication efficiency, perception accuracy, system response speed, and the safety of real-time applications. Higher transmission frequencies can provide more timely and rich sensory information. This also implies higher consumption of communication resources. However, in a dynamic and complex environment, it is difficult to quantitatively control the message transmission frequency so as to improve the utilization of limited communication resources, ensure the timeliness of perception messages, and maintain fast convergence. To address challenges of timeliness of messages, a timeliness performance metric age of information (AoI) is introduced to control message transmission frequency. This paper deduced the average AoI of the system and designed an AoI-aware decentralized congestion control (ADCC) algorithm for the V2X-based cooperative perception system. Simulation results show that the ADCC algorithm outperforms the classic congestion control algorithm linear adaptive message rate (LIMERIC) in terms of channel utilization and throughput. Specifically, AoI decreased by 47.3%, channel utilization increased by 6.5% and throughput increased by 48.6%.

Network function virtualization (NFV) enables the provision of different quality of service (QoS) levels through service function chains (SFCs), where NFV outsources big data tasks of end users to nearby edge servers. In multi-access edge computing (MEC), its dynamic and uncertainty nature poses great challenges to the SFC placement problem, which requires optimizing multiple potentially-conflicting objectives, such as network latency and load balancing. Moreover, user preferences may vary along with time, adding another layer of complexity to the problem. To address the problem above, we propose a novel distributed deep reinforcement learning (DRL) architecture based on a spatio-temporal encoder (STE), denoted as DDRL-STE. DDRL-STE is featured with equal-weight pre-training and transformer-based STE. Experimental results show that DDRL-STE outperforms three state-of-the-art DRL algorithms regarding latency and load balancing under three well-known network topologies, exhibiting its excellent potential in exploration and generalization.

In this work, we propose TARS, a Topology-aware AllReduce algorithm Synthesizer, to generate optimal execution plans for AllReduce workloads over arbitrary interconnection network structures. Distinguished from existing topology-aware synthesizers that formulate the two stages of AllReduce (e.g., ReduceScatter-then-AllGather, or Reduce-then-Broadcast) separately, the power of TARS stems from employing a comprehensive Integer Quadratic Programming (IQP) model to formulate the entire workflow precisely. Preliminary studies confirm that, compared with the state-of-the-art scheme, TARS could significantly reduce the completion time of AllReduce.

Mobile edge computing has been a promising solution to enable real-time service in vehicular networks. However, due to high dynamics of mobile environment and heterogeneous features of vehicular services, traditional expert-based or learning-based strategies has to update handcrafted parameters or retrain learning model, which leads to intolerant overhead. Therefore, this paper investigates the problem of multi-task offloading (MTO), where there exist multiple offloading scenarios with varying parameters, such as task topology, resource requirement and transmission/computation capability. The objective is to design a unified solution to minimize task execution time under different MTO scenarios. Accordingly, we develop a Seq2seq-based Meta Reinforcement Learning algorithm for MTO (SMRL-MTO). Specifically, a bidirectional gated recurrent units integrated with attention mechanism is designed to determine offloading action by encoding sequential offloading actions and showing different preferences to different parts of input sequence. Particularly, a meta reinforcement learning framework is designed based on model-agnostic meta learning, which trains a meta policy offline and fast adapts to new MTO scenario within a few training steps. Finally, we conduct performance evaluation based on task generator DAGGEN and realistic vehicular traces, which shows that the SMRL-MTO reduces task execution time by 11.36% on average compared with greedy algorithm.

Mobile edge computing (MEC) is expected to support real-time services at wireless networks, where task replication is applied to guarantee job completion within a strict deadline through replicating multiple copies to different edge servers. Most of previous works focused on guaranteeing the reliability of individual task in MEC-based networks with the assumption of homogeneous task execution distribution. Further, these algorithms cannot suit dynamic network scales, due to overhigh communication or retraining overhead. Therefore, this paper formulates the problem of heterogeneous task replication in a finer level by modeling outage probability of individual replication, where the decisions of all tasks are jointly optimized within the constraints of both mobile users and MEC servers for minimizing job outage probability. To adapt to varying network scales, we develop centralized and distributed algorithms, respectively. The centralized algorithm is developed based on Interior Point Method, which obtains the optimal solution of relaxed model and then approximates to the solution of original problem. Further, the distributed algorithm decomposes the HTR into multiple subproblems and parallelly compute each local solution based on Distributed ADMM. Finally, we build a simulation model and conduct comprehensive results, which demonstrates that the proposed algorithms can achieve high-accuracy solution with fast convergence.

Multi-user semantic communication (MUSC) has emerged as a promising paradigm for future 6G networks and applications, where massive clients (e.g., mobile devices) collaboratively construct a global semantic decoder without sharing their local data. However, due to the lack of direct access to clients’ data, MUSC is vulnerable to data poisoning attacks (DPAs), wherein malicious participants send updates derived from poisoned training samples. Current defense techniques against DPAs are designed for traditional networks and are not directly applicable to MUSC. In this paper, we propose an effective attack-defense game framework, denoted as DPAD-MUSC, tailored to defend against DPAs during image transmission for MUSC. First, we determine each attack-type's optimal attack policy based on reinforcement learning, with the aim of strengthening the attack while avoiding detection. To generate adversarial samples accordingly, we devise an adversarial samples generator (ADV-Generator) based on conditional generative adversarial network (CGAN). Then, we introduce an attack defender (DPA-Defender) to detect data poisoning attacks and exclude poisoned samples from the target model's learning process, with the adversarial samples generated under the guidance of the optimal attack policy to enhance the detector's robustness. Simulation results demonstrate that the DPAD-MUSC can find optimal attack policies that cause a greater accuracy drop in the target model while maintaining a higher evasion rate. The ADV-Generator can generate effective adversarial samples and the DPA-Defender outperforms five state-of-the-art methods on three widely used image datasets under additive white Gaussian noise (AWGN) channel in terms of Top-1 accuracy.

This paper investigates the problem of energy-efficient trajectory optimization with wireless charging (ETWC) in an unmanned aerial vehicle (UAV)-assisted mobile edge computing system. A UAV is dispatched to collect computation tasks from specific ground smart devices (GSDs) within its coverage while transmitting energy to the other GSDs. In addition, a high-altitude platform with a laser beam is deployed in the stratosphere to charge the UAV, so as to maintain its flight mission. The ETWC problem is characterized by multi-objective optimization, aiming to maximize both the energy efficiency of the UAV and the number of tasks collected via optimizing the UAV's flight trajectories. The conflict between the two objectives in the problem makes it quite challenging. Recently, some single-objective reinforcement learning (SORL) algorithms have been introduced to address the aforementioned problem. Nevertheless, these SORLs adopt linear scalarization to define the user utility, thus ignoring the conflict between objectives. Furthermore, in dynamic MEC scenarios, the relative importance assigned to each objective may vary over time, posing significant challenges for conventional SORLs. To solve the challenge, we first build a multi-objective Markov decision process that has a vectorial reward mechanism. There is a corresponding relationship between each component of the reward and one of the two objectives. Then, we propose a new trace-based experience replay scheme to modify sample efficiency and reduce replay buffer bias, resulting in a modified multi-objective reinforcement learning algorithm. The experiment results validate that the proposed algorithm can obtain better adaptability to dynamic preferences and a more favorable balance between objectives compared with several algorithms.

This paper studies the age of information (AoI) and energy tradeoff (AET) problem in an aerial-ground collaborative mobile edge computing system, where a high-altitude platform and an unmanned aerial vehicle (UAV) work together to offer computing services for ground devices (GDs). The AET problem is formulated as a multi-objective optimization problem (MOP) that aims at simultaneously minimizing the total AoI of GDs and total energy consumption of the UAV by optimizing its flight paths and task offloading ratios. Addressing the AET problem poses a significant challenge due to the inherent conflict between the two objectives. The existing methods cannot well address the MOP because they adopt the linear combination to transform an MOP into a single-objective optimization problem using fixed weights (i.e., preferences), ignoring the conflict between objectives. Moreover, user preferences may change over time in dynamic MEC systems. To overcome these challenges, we first build a multi-objective Markov decision process model with a vectorial reward for the AET problem. There are one-to-one relationships between each component of the reward and one of the two objectives. Then, we propose a multi-objective learning algorithm based on proximal policy optimization (PPO), which primarily comprises a training phase and an evolutionary phase. The former adopts multi-objective PPO to iteratively optimize multiple learning individuals, aiming to obtain a nondominated policy set. The latter employs a genetic operator to further improve the quality of each policy in the set. Specifically, the crossover and mutation operators operate at the parameter level of policy networks, avoiding stagnation and premature convergence. The experiment results validate that the proposed approach obtains a set of excellent nondominated policies and a favorable balance between objectives. Moreover, the proposed approach achieves improvements of at least 39.8%, 2.1%, and 15.3% regarding AoI, energy consumption, and cost compared with several algorithms.

Time series data widely exist in public services, industrial environments, and military applications. Traditionally, the transmission of a huge volume of data for analytic tasks poses challenges, particularly in mobile environments with limited computing and communication resources. Semantic communication emerges as a solution for intelligently extracting various features from source data and efficiently transmitting task-related information to receivers, thereby reducing bandwidth consumption significantly. In this paper, we introduce a novel federated semantic communication system tailored for forecasting-oriented time series transmission tasks. The correlation of source data collected from terminal devices is mined and the corresponding semantic information is transmitted to an edge server for collaborative inference. To optimize the semantic analysis process, we devise a deep decomposition block at the transmitter side, decomposing time series into trend and multiple period components. This reduces noise interference from wireless channels, enhancing the overall transmission quality. For effective training and collaborative inference, we propose a Federated Mixture of period Routers (FedMoR) architecture. Within each channel encoder, period routers are divided into private and public ones. Private routers extract specialized features from individually collected data, mitigating accuracy degradation. Public routers share knowledge across all transmitters, enhancing temporal analysis robustness. Simulation results demonstrate that the proposed system outperforms two traditional technique-based and two semantic communication-based baselines under three common channels. The system achieves low mean square errors on five widely-used real-world time series forecasting datasets, particularly in the low signal-to-noise ratio regime.

By offloading partial of the aggregation computation from the logical central parameter servers to network devices like programmable switches, In-Network Aggregation (INA) is a general, effective, and widely used approach to reduce network load thus alleviating the communication bottlenecks suffered by large-scale distributed training. Given the fact that INA would take effects if and only if associated traffic goes through the same in-network aggregator, the key to taking advantage of INA lies in routing control. However, existing proposals fall short in doing so and thus are far from optimal, since they select routes for INA-supported traffic without comprehensively considering the characteristics, limitations, and requirements of the network environment, aggregator hardware, and distributed training jobs. To fill the gap, in this paper, we systematically establish a mathematical model to formulate i) the up-down routing constraints of Clos datacenter networks, ii) the limitations raised by modern programmable switches’ pipeline hardware structure, and iii) the various aggregator-aware routing optimization goals required by distributed training tasks under different parallelism strategies. Based on the model, we develop ARO, an Aggregator-aware Routing Optimization solution for INA-accelerated distributed training applications. To be efficient, ARO involves a suite of search space pruning designs, by using the model’s characteristics, yielding tens of times improvement in the solving time with trivial performance loss. Extensive experiments show that ARO is able to find near-optimal results for large-scale routing optimization in tens of seconds, achieving$1.8\sim 4.0\times $higher throughput than the state-of-the-art solution.

By allowing$p$out of$n$workers to conductall reduceoperations among them for a round of synchronization,partial reduce, a promising partially-asynchronous variant ofall reduce, has shown its power in alleviating the impacts of stragglers for iterative distributed machine learning (DML). Currentpartial reducesolutions are mainly designed for intra-cluster DML, in which workers are networked with high-bandwidth LAN links. Yet no prior work has looked into the problem of how to achieve efficientpartial reducefor cross-cloud DML, where inter-worker connections are with scarcely-available capacities. To fill the gap, in this paper, we proposeCREW, a flexible and efficient implementation ofpartial reducefor cross-cloud DML. At the high level,CREWis built upon the novel design of employing all active workers along with their internal connection capacities to execute the involved communication and computation tasks; and at the low level,CREWemploys a suite of algorithms to distribute the tasks among workers in a load-balanced way, and deal with possible outages of workers/connections, and bandwidth contention. Detailed performance studies confirm that,CREWnot only shortens the execution of eachpartial reduceoperation, outperforming existing communication schemes such as PS, Ring,TopoAdopt, and BLINK greatly, but also significantly accelerates the training of large models, up to$15\times$and$9\times$, respectively, when compared with the all-to-all direct communication scheme andoriginal partial reducedesign.

Multivariate time series classification (MTSC) based on deep learning (DL) has attracted increasingly more research attention. The performance of a DL-based MTSC algorithm is heavily dependent on the quality of the learned representations providing semantic information for downstream tasks, e.g., classification. Hence, a model’s representation learning ability is critical for enhancing its performance. This article proposes a densely knowledge-aware network (DKN) for MTSC. The DKN’s feature extractor consists of a residual multihead convolutional network (ResMulti) and a transformer-based network (Trans), called ResMulti-Trans. ResMulti has five residual multihead blocks for capturing the local patterns of data while Trans has three transformer blocks for extracting the global patterns of data. Besides, to enable dense mutual supervision between lower-and higher-level semantic information, this article adapts densely dual self-distillation (DDSD) for mining rich regularizations and relationships hidden in the data. Experimental results show that compared with 5 state-of-the-art self-distillation variants, the proposed DDSD obtains 13/4/13 in terms of “win”/“tie”/“lose” and gains the lowest-AVG_rank score. In particular, compared with pure ResMulti-Trans, DKN results in 20/1/9 regarding win/tie/lose. Last but not least, DKN overweighs 18 existing MTSC algorithms on 10 UEA2018 datasets and achieves the lowest-AVG_rank score.

Health systems worldwide have an unprecedented opportunity to enhance healthcare service delivery due to the rapid development of emerging digital technologies. Many advancements have been made in the medical field, with deep learning proving particularly useful when applied to a large enough number of well-defined samples. Although, this aspect may make deep learning harder to implement in settings with limited-size datasets. In this study, we present a new method of chronic kidney disease detection (CKDD) by combining Generative Adversarial Networks (GAN) with Convolutional Neural Networks (CNN). Afterward, synthetic sample data was created using GAN, which enlarged the dataset. Subsequently, processing these synthetic samples, the CNN classifier was applied. According to experimental assessments, the suggested CKDD-GAN methodology accuracy is superior to without the GAN technique. Moreover, the proposed CKDD-GAN-based model outperformed with an accuracy of 98.10%. Even though standard synthetic data samples seemed to improve classification performance, GAN-based enhancements resulted in a 2.91% improvement. GAN implementations for detecting chronic kidney disease are highly beneficial since they also increase awareness about its possible uses in various other diseases.

This paper proposes a task-oriented semantic communication system for electrocardiogram (ECG) signal classification, called ECG-SC-DARTS. Based on deep learning, this system adopts the differentiable neural architecture search (DARTS) to automatically design the neural architecture of the semantic encoder under various channels. This paper improves the performance of the original DARTS by introducing a new recurrent neural network (RNN) cell with residual structure and a noise adding scheme for skip-connections. The RNN cell enhances the temporal semantic information extraction ability while the added noise reduces the risk of performance collapse caused by skip-connections. Experimental results demonstrate that ECGSC-DARTS generates appropriate neural architectures for the semantic encoder under AWGN, Rayleigh and Rician channels and these architectures outperform a number of baseline models, such as the original DARTS, fully convolutional network, multi-layer perception, and ResNet, regarding F1-score. Moreover, ECGSC-DARTS is more reliable than the traditional communication system in harsh channel environment.

We present a many-to-one distributed semantic communication system for multivariate time series classification. The system adopts a federated learning-based architecture to achieve low-redundancy collaborative inference, where an unsupervised auxiliary task is designed to coordinate the feature vectors at different dimensions between semantic encoders and the classifier. For each transmitter, we design a scale-adaptive semantic encoder by applying weighted sum to a number of predefined convolutional layers. The scale-adaptive semantic encoder can extract multi-scale features from time series following various distributions. A dynamic channel encoder is developed to adapt to the scale-adaptive semantic encoder, converting semantic features to complex symbols appropriate for wireless transmission. For the receiver, we apply the same scale-adaptive structure to the semantic decoder to extract multi-scale semantic features from all transmitters for accurate classification. Simulation results show that the proposed distributed semantic communication system outperforms two baseline systems under AWGN, Rician, and Rayleigh channels and achieves excellent Top-1 accuracy performance on three UEA2018 datasets, especially when the signal-to-noise ratio is low.

The multipoint expected improvement (EI) criterion is a well-defined parallel infill criterion for expensive optimization. However, the exact calculation of the classical multipoint EI involves evaluating a significant amount of multivariate normal cumulative distribution functions, which makes the inner optimization of this infill criterion very time consuming when the number of infill samples is large. To tackle this problem, we propose a novel fast multipoint EI criterion in this work. The proposed infill criterion is calculated using only univariate normal cumulative distributions; thus, it is easier to implement and cheaper to compute than the classical multipoint EI criterion. It is shown that the computational time of the proposed fast multipoint EI is several orders lower than the classical multipoint EI on the benchmark problems. In addition, we propose to use cooperative coevolutionary algorithms (CCEAs) to solve the inner optimization problem of the proposed fast multipoint EI by decomposing the optimization problem into multiple subproblems with each subproblem corresponding to one infill sample and solving these subproblems cooperatively. Numerical experiments show that using CCEAs can improve the performance of the proposed algorithm significantly compared with using standard evolutionary algorithms. This work provides a fast and efficient approach for parallel expensive optimization.

This paper studies the trajectory control and task offloading (TCTO) problem in an unmanned aerial vehicle (UAV)-assisted mobile edge computing system, where a UAV flies along a planned trajectory to collect computation tasks from smart devices (SDs). We consider a scenario that SDs are not directly connected by the base station (BS) and the UAV has two roles to play: MEC server or wireless relay. The UAV makes task offloading decisions online, in which the collected tasks can be executed locally on the UAV or offloaded to the BS for remote processing. The TCTO problem involves multi-objective optimization as its objectives are to minimize the task delay and the UAV's energy consumption, and maximize the number of tasks collected by the UAV, simultaneously. This problem is challenging because the three objectives conflict with each other. The existing reinforcement learning (RL) algorithms, either single-objective RLs or single-policy multi-objective RLs, cannot well address the problem since they cannot output multiple policies for various preferences (i.e. weights) across objectives in a single run. An evolutionary multi-objective RL (EMORL) algorithm is applied to address the TCTO problem. We improve the multi-task multi-objective proximal policy optimization of the original EMORL by retaining all new learning tasks in the offspring population, which can preserve promissing learning tasks. The simulation results demonstrate that the proposed algorithm can obtain more excellent non-dominated policies by striking a balance between the three objectives regarding policy quality, compared with two evolutionary algorithms, two multi-policy RL algorithms, and the original EMORL.

Recently, the abstraction ofcoflowis introduced to capture the collective data transmission patterns among modern distributed data-parallel applications. During processing, coflows generally act as barriers; accordingly, time-sensitive applications prefer their coflows to complete within deadlines, and deadline-aware coflow scheduling becomes very crucial. Regarding these data-parallel applications, we notice that many of them, includinglarge-scale query systems,distributed iterative training, anderasure codes enabled storage, are able to tolerate loss-bounded incomplete inputs by design. This tolerance indeed brings a flexible design space for the schedule of their coflows: when getting overloaded, the network can trade coflow completeness for the timeliness, and balance the completeness of different coflows on demand. Unfortunately, existing coflow schedulers neglect this tolerance, resulting in inflexible and inefficient bandwidth allocations. In this paper, we explore this fundamental trade-off and design POCO, a POlicy-based COflow scheduler, along with a transport layer enhancement scheme, to achieve customizable selective coflow completion for emerging time-sensitive distributed applications. Internally, POCO employs a suite of novel designs along with admission controls to makeflexible,work-conserving, andperformance-guaranteedrate allocation to online coflow requests very efficiently. Extensive trace-based simulations indicate that POCO is highly flexible and achieves optimal coflow schedules respecting the requirements specified by applications.

Multi-access edge computing (MEC) and network function virtualization (NFV) are promising technologies to support emerging IoT applications, especially those computation-intensive. In NFV-enabled MEC environment, service function chain (SFC), i.e., a set of ordered virtual network functions (VNFs), can be mapped on MEC servers. Mobile devices (MDs) can offload computation-intensive applications, which can be represented by SFCs, fully or partially to MEC servers for remote execution. This article studies the partial offloading and SFC mapping joint optimization (POSMJO) problem in an NFV-enabled MEC system, where the data from an incoming task is partitioned into two parts, with one part executed locally and the other offloaded to the edge infrastructure for execution. These two parts are independent of each other, but both need to be processed by the same SFC. The objective is to minimize the average cost in the long term which is a combination of execution delay, MD's energy consumption, and usage charge for edge computing. This problem consists of two closely related decision-making steps, namely task partition and VNF placement, which is highly complex and quite challenging. To address this, we propose a cooperative dual-agent deep reinforcement learning (CDADRL) algorithm, where two agents interact with each other. Simulation results show that the proposed algorithm outperforms three combinations of deep reinforcement learning algorithms with respect to cumulative reward and it overweighs a number of baseline algorithms in terms of execution delay, energy consumption, and usage charge.

No abstract available.

No abstract available.

Recent advances in distributed machine learning show theoretically and empirically that, for many models, provided workers would participate in the synchronizations eventually, i) the training still converges, even if only p workers take part in each round of synchronization, and ii) a larger p generally leads to a faster rate of convergence. These findings shed light on eliminating the bottleneck effects of parameter synchronization in large-scale data-parallel distributed training, having motivated several optimization designs.In this paper, we focus on optimizing the parameter synchronization for peer-to-peer distributed learning, in which workers generally broadcast or multicast their updated parameters to others for synchronization, and propose SELMCAST, an expressive and Pareto-optimal multicast receiver selection algorithm, to achieve the goal. Compared with the state-of-the-art design that randomly selects exactly p receivers for each worker’s multicast in a bandwidth-agnostic way, SELMCAST chooses receivers based on the global view of their available bandwidth and loads, yielding two advantages. Firstly, it could optimize the bottleneck sending rate, thus cutting down the time cost of parameter synchronization. Secondly, when more than p receivers are with sufficient bandwidth, they would be selected as many as possible, bringing benefits to the convergence of training. Extensive evaluations show that SELMCAST is efficient and always achieves near-optimal performance.

Nowadays, cross-device federated learning (FL) is the key to achieving personalization services for mobile users and has been widely employed by companies like Google, Microsoft, and Alibaba in production. With the explosive increase of participants, the central FL server, which acts as the manager and aggregator of cross-device model training, would get overloaded, becoming the system bottlenecks. Inspired by the emerging wave of edge computing, an interesting question is: could edge clouds help cross-device FL systems overcome the bottleneck?This article provides a cautiously optimistic answer by proposing INP, an FL-specific In-Network Processing framework, along with the novel Model Download Protocol of MDP and Model Upload Protocol of MUP. With MDP and MUP, edge cloud nodes along the paths in INP can easily eliminate duplicated model downloads and pre-aggregate associated gradient uploads for the central FL server, thus alleviating its bottleneck effect, and further accelerating the entire training progress significantly.

To improve the performance of partial reduce on synchronizing models for heterogeneous data-parallel distributed training, we explore the idea of selective waiting and worker selection to propose the flexible solution of selective reduce. Our preliminary study shows that with progress- and bandwidth-aware decisions, the proposed partial reduce outperforms the original partial reduce significantly, in terms of both the average synchronization scales and completion times.

Over the years, a number of semisupervised deep-learning algorithms have been proposed for time-series classification (TSC). In semisupervised deep learning, from the point of view of representation hierarchy, semantic information extracted from lower levels is the basis of that extracted from higher levels. The authors wonder if high-level semantic information extracted is also helpful for capturing low-level semantic information. This paper studies this problem and proposes a robust semisupervised model with self-distillation (SD) that simplifies existing semisupervised learning (SSL) techniques for TSC, called SelfMatch. SelfMatch hybridizes supervised learning, unsupervised learning, and SD. In unsupervised learning, SelfMatch applies pseudolabeling to feature extraction on labeled data. A weakly augmented sequence is used as a target to guide the prediction of a Timecut-augmented version of the same sequence. SD promotes the knowledge flow from higher to lower levels, guiding the extraction of low-level semantic information. This paper designs a feature extractor for TSC, called ResNet–LSTMaN, responsible for feature and relation extraction. The experimental results show that SelfMatch achieves excellent SSL performance on 35 widely adopted UCR2018 data sets, compared with a number of state-of-the-art semisupervised and supervised algorithms.

Mobile edge computing (MEC) has been an effective paradigm for supporting computation-intensive applications by offloading resources at network edge. Especially in vehicular networks, the MEC server, is deployed as a small-scale computation server at the roadside and offloads computation-intensive task to its local server. However, due to the unique characteristics of vehicular networks, including high mobility of vehicles, dynamic distribution of vehicle densities and heterogeneous capacities of MEC servers, it is still challenging to implement efficient computation offloading mechanism in MEC-assisted vehicular networks. In this article, we investigate a novel scenario of computation offloading in MEC-assisted architecture, where task upload coordination between multiple vehicles, task migration between MEC/cloud servers and heterogeneous computation capabilities of MEC/cloud severs, are comprehensively investigated. On this basis, we formulate cooperative computation offloading (CCO) problem by modeling the procedure of task upload, migration and computation based on queuing theory, which aims at minimizing the delay of task completion. To tackle the CCO problem, we propose a probabilistic computation offloading (PCO) algorithm, which enables MEC server to independently make online scheduling based on the derived allocation probability. Specifically, the PCO transforms the objective function into augmented Lagrangian and achieves the optimal solution in an iterative way, based on a convex framework called Alternating Direction Method of Multipliers (ADMM). Last but not the least, we implement the simulation model. The comprehensive simulation results show the superiority of the proposed algorithm under a wide range of scenarios.

Fog computing has been an effective paradigm of real-time applications in the IoT area, which enables task offloading at network edge devices. Particularly, many emerging vehicular applications require real-time interaction between the terminal users and computation servers, which can be implemented in fog-based architecture. However, it is still challenging to apply fog computing in vehicular networks due to high mobility of vehicles and uneven distribution of vehicle density, which may result in performance degradation, such as unbalanced workload and unexpected task failure. In this article, we investigate a new service scenario of task offloading under a three-layer service architecture, where the resources of vehicular fog (VF), fog server (FS), and central cloud (CC) are utilized in a cooperative way. On this basis, we formulate the probabilistic task offloading (PTO) problem by synthesizing task transmission, computation, and result retrieval, as well as characterizing the heterogeneity of computation servers. The objective of the PTO is to minimize the weighted sum of execution delay, energy consumption, and payment cost. To resolve the PTO problem, we propose a comprehensive task offloading algorithm by combining the alternating direction method of multipliers (ADMMs) and particle swarm optimization (PSO), called ADMM-PSO. The basic idea of the ADMM-PSO is to divide the PTO problem into multiple unconstrained subproblems and achieve the optimal solution in the form of an iterative coordination process. For each iteration, the solution is achieved by solving each subproblem with the PSO and updated based on a designed rule, which is able to converge to the optimal solution when the stop criterion is satisfied. Finally, we build the simulation model and implement the proposed algorithm for performance evaluation. The simulation results demonstrate the superiority of the proposed algorithm under a wide range of service scenarios.

Prescreening strategies have been widely used in surrogate-assisted evolutionary algorithms for screening out poor solutions. Existing prescreening strategies are designed for individual-level selection, i.e. they are used to select individuals from a set of population members. In this work, we propose a population prescreening strategy based on the multi-point expected improvement criterion for Kriging-assisted evolutionary algorithms. In each generation of the proposed algorithm, the evolutionary operators are used repeatedly to generate a set of candidate populations. Then, these candidate populations are prescreened by the multi-point expected improvement criterion and the population with highest multi-point expected improvement value is selected for the next generation. Following this, infill criteria are used to select promising solutions from the selected population for expensive evaluation. Numerical experiments on eighteen test problems show that the proposed population prescreening strategy can improve the optimization efficiency of the Kriging-assisted evolutionary algorithms significantly without introducing too much additional computational cost.

Over the years, a large number of deep learning algorithms have been developed for time series classification (TSC). These algorithms were usually invented by researchers with prior knowledge and experience. However, it is a critical challenge for beginners to design decent structures to address various TSC problems. To this end, we propose a robust neural temporal search (RNTS) framework for identifying the relationships and features in TSC data, which mainly contains a temporal search network and an attentional LSTM network. To be specific, inspired by the idea of neural architecture search (NAS), the temporal search network automatically transforms its structure for each dataset according to its characteristics, responsible for extracting basic features. The attentional LSTM network is used to explore the complex shapelets and relationships the former may ignore. Experimental results demonstrate that RNTS achieves the best overall performance on 24 standard datasets selected from the UCR 2018 archive, in terms of three measures based on the top-l accuracy, compared with a number of state-of-the-art approaches.

Mobile edge computing (MEC) has been an effective paradigm to support real-time computation-intensive vehicular applications. However, due to highly dynamic vehicular topology, these existing centralized-based or distributed-based scheduling algorithms requiring high communication overhead, are not suitable for task offloading in vehicular networks. Therefore, we investigate a novel service scenario of MEC-based vehicular crowdsourcing, where each MEC server is an independent agent and responsible for making scheduling of processing traffic data sensed by crowdsourcing vehicles. On this basis, we formulate a data-driven task offloading problem by jointly optimizing offloading decision and bandwidth/computation resource allocation, and renting cost of heterogeneous servers, such as powerful vehicles, MEC servers and cloud, which is a mixed-integer programming problem and NP-hard. To reduce high time-complexity, we propose the solution in two stages. First, we design an asynchronous deep Q-learning to determine offloading decision, which achieves fast convergence by training the local DQN model at each agent in parallel and uploading for global model update asynchronously. Second, we decompose the remaining resource allocation problem into several independent subproblems and derive optimal analytic formula based on convex theory. Lastly, we build a simulation model and conduct comprehensive simulation, which demonstrates the superiority of the proposed algorithm.

This article studies the joint optimization problem of computation offloading and resource allocation (JCORA) in mobile-edge computing (MEC). Deep reinforcement learning (DRL) is one of the ideal techniques for addressing the dynamic JCORA problem. However, it is still challenging to adapt traditional DRL methods for the problem since they usually lead to slow and unstable convergence in model training. To this end, we propose a temporal attentional deterministic policy gradient (TADPG) to tackle JCORA. Based on the deep deterministic policy gradient (DDPG), TADPG has two significant features. First, a temporal feature extraction network consisting of a 1-D convolution (Conv1D) residual block and an attentional long short-term memory (LSTM) network is designed, which is beneficial to high-quality state representation and function approximation. Second, a rank-based prioritized experience replay (rPER) method is devised to accelerate and stabilize the convergence of model training. Experimental results demonstrate that the decentralized TADPG-based mechanism can achieve more efficient JCORA performance than the centralized one, and the proposed TADPG outperforms a number of state-of-the-art DRL agents in terms of the task completion time and energy consumption.

Kriging models, also known as Gaussian process models, are widely used in surrogate-assisted evolutionary algorithms (SAEAs). However, the cubic time complexity of the standard Kriging models limits their usage in high-dimensional optimization. To tackle this problem, we propose an incremental Kriging model for high-dimensional surrogate-assisted evolutionary computation. The main idea is to update the Kriging model incrementally based on the equations of the previously trained model instead of building the model from scratch when new samples arrive, so that the time complexity of updating the Kriging models can be reduced to quadratic. The proposed incremental learning scheme is very suitable for online SAEAs since they evaluate new samples in each one or several generations. The proposed algorithm is able to achieve competitive optimization results on the test problems compared with the standard Kriging-assisted evolutionary algorithm and is significantly faster than the standard Kriging approach. The proposed algorithm also shows competitive or better performances compared with four fast Kriging-assisted evolutionary algorithms and four state-of-the-art SAEAs. This work provides a fast way of employing Kriging models in high-dimensional surrogate-assisted evolutionary computation.

Dynamic routing in software-defined networking (SDN) can be viewed as a centralized decision-making problem. Most of the existing deep reinforcement learning (DRL) agents can address it, thanks to the deep neural network (DNN) incorporated. However, fully-connected feed-forward neural network (FFNN) is usually adopted, where spatial correlation and temporal variation of traffic flows are ignored. This drawback usually leads to significantly high computational complexity due to large number of training parameters. To overcome this problem, we propose a novel model-free framework for dynamic routing in SDN, which is referred to as spatio-temporal deterministic policy gradient (STDPG) agent. Both the actor and critic networks are based on identical DNN structure, where a combination of convolutional neural network (CNN) and long short-term memory network (LSTM) with temporal attention mechanism, CNN-LSTM-TAM, is devised. By efficiently exploiting spatial and temporal features, CNN-LSTM-TAM helps the STDPG agent learn better from the experience transitions. Furthermore, we employ the prioritized experience replay (PER) method to accelerate the convergence of model training. The experimental results show that STDPG can automatically adapt for current network environment and achieve robust convergence. Compared with a number state-of the-art DRL agents, STDPG achieves better routing solutions in terms of the average end-to-end delay.

Recently, the abstraction of coflow is introduced to capture the collective data transmission patterns among modern distributed data-parallel application. During processing, coflows generally act as barriers; accordingly, time-sensitive applications prefer their coflows to complete within deadlines and deadline-aware coflow scheduling becomes very crucial.

As the key infrastructure for emerging 5G and Internet-of-Things (IoT) applications, micro data centers would be widely deployed at network edges to provide high-bandwidth low-latency cloud service. In these systems, applications would deliver large-size data objects among servers for various purposes like service deployment, application scale-up, and data duplication on demand. Accordingly, reducing delivery time is crucial for the optimization of service delay and system utilization. To accelerate the delivery, this article proposes a multisource-aware adaptive data transmission solution, Parallel Push (PPUSH), by leveraging the fact that data objects in the cloud are generally replicated among servers by design. At the high level, PPUSH achieves efficient delivery of multisource data by launching multiple push flows in parallel; and at the low level, it decouples transfers from different sources by encoding data objects with rateless RaptorQ code, and further employing novel congestion controls to prioritize the bandwidth allocation of concurrent tasks respecting their remaining sizes. Fluid model analysis along with Mininet-based test and packet-level simulation shows that, unlike DCTCP and other proposals, push is robust to packet loss and achieves provable prioritized bandwidth allocation. Extensive simulation results imply that, with above advantages, PPUSH could achieve very efficient data delivery by making use of all available data sources: for instance, compared with the straightforward design of equal-size task split and fair bandwidth allocation, its adaptive task assignment and prioritized traffic scheduling reduce the average task completion time in a tested scenario by 1.495× and 1.329×, respectively, demonstrating a total improvement of 1.586×, when enabled at the same time.

In mobile-edge computing (MEC), smart mobile devices (SMDs) with limited computation resources and battery lifetime can offload their computing-intensive tasks to MEC servers, thus to enhance the computing capability and reduce the energy consumption of SMDs. Nevertheless, offloading tasks to the edge incurs additional transmission time and thus higher execution delay. This article studies the tradeoff between the completion time of applications and the energy consumption of SMDs in MEC networks. The problem is formulated as a multiobjective computation offloading problem (MCOP), where the task precedence, i.e., ordering of tasks in SMD applications, is introduced as a new constraint in the MCOP. An improved multiobjective evolutionary algorithm based on decomposition (MOEA/D) with two performance enhancing schemes is proposed: 1) the problem-specific population initialization scheme uses a latency-based execution location (EL) initialization method to initialize the EL (i.e., either local SMD or MEC server) for each task and 2) the dynamic voltage and frequency scaling-based energy conservation scheme helps to decrease the energy consumption without increasing the completion time of applications. The simulation results clearly demonstrate that the proposed algorithm outperforms a number of state-of-the-art heuristics and metaheuristics in terms of the convergence and diversity of the obtained nondominated solutions.

In today's data center networks (DCN), cloud applications commonly disseminate files from a single source to a group of receivers for service deployment, data replication, software upgrade, etc. For these group communication tasks, recent advantages of software-defined networking (SDN) provide bandwidth-efficient ways-they enable DCN to establish and control a large number of explicit multicast trees on demand. Yet, the benefits of data center multicast are severely limited, since there does not exist a scheme that could prioritize multicast transfers respecting the performance metrics wanted by today's cloud applications, such as pursuing small mean completion times or meeting soft-time deadlines with high probability. To this end, we propose PAM (Priority-based Adaptive Multicast), a preemptive, decentralized, and ready-deployable rate control protocol for data center multicast. At the core, switches in PAM explicitly control the sending rates of concurrent multicast transfers based on their desired priorities and the available link bandwidth. With different policies of priority generation, PAM supports a range of scheduling goals. We not only prototype PAM upon the emerged P4-based programmable switch with novel approximation designs, but also evaluate its performance with ns3-based extensive simulations. Results imply that PAM is ready-deployable; it converges very fast, has negligible impacts on coexisting TCP traffic, and always performs near-optimal priority-based multicast scheduling.

Mobile edge computing (MEC) has been an emerging paradigm to support low-latency applications in vehicular networks by offloading resources at network edge. However, it is still challenging to apply MEC- based architecture to implement multimedia services due to varying wireless communication, high vehicle mobility and heterogeneous resource integration. In this paper, we investigate adaptive-bitrate (ABR)-based multimedia services (MS) in MEC-based vehicular networks, where each multimedia file is divided into multiple chunks and can be requested at different bitrate levels. Further, MEC servers can satisfy local vehicular requests by integrating heterogeneous cache and communication resources. Based on the above observation, we formulate joint resource optimization (JSO) problem by synthesizing cache placement, wireless bandwidth allocation and chunk quality adaptation. On this basis, we propose a reinforcement- learning-based cache placement (RLCP) algorithm, which determines the optimal offloaded chunks by learning the global knowledge of cache reward in an iterative way. Further, we design an adaptive-quality- based chunk selection (AQCS) algorithm, which can be adaptive to time-varying wireless channel by dynamically adjusting bandwidth allocation and quality level based on real-time service workload. Lastly, we build the simulation model and conduct an extensive performance evaluation, which demonstrates the superiority of proposed algorithms.

Mobile edge computing (MEC) is an emerging paradigm to offload the server-side resources closer to the mobile terminals compared with cloud-based computing. However, due to highly vehicular mobility and limited wireless coverage, it is challenging to apply off-the-shelf MEC-based architecture to support the real-time services in vehicular networks, especially when the vehicle density changes dynamically. Hence, this paper investigates a novel service scenario in an MEC-based architecture, where the local MEC server has to complete the real-time services of mobile vehicles in its service range. On this basis, we formulate a novel problem of distributed real-time service scheduling (DRSS) by comprehensively considering the delay requirements of real-time services, the heterogeneous computing capabilities of MEC servers and the mobility features of vehicles, which targets at maximizing the service ratio. To resolve such an issue, we propose a multi-agent reinforcement learning algorithm called Utility-based Learning (UL), in which each local MEC server selects the optimal solution by learning the global knowledge online. Specifically, a utility table is established to determine the optimal solution by estimating the pending delay of service request at each MEC server and it will be updated periodically based on the feedback signal from the assigned MEC server. Lastly, we build the simulation model and conduct an extensive performance evaluation, which demonstrates the superiority of the proposed algorithm.

Heterogeneous network integration is a promising technique to support efficient data services in vehicular networks. However, due to highly dynamics of vehicular mobility and heterogeneous performance of wireless interfaces, it is still challenging to design an efficient scheduling policy for information services in vehicular networks. In this paper, we propose a centralized service architecture for managing heterogeneous network resources. Particularly, we comprehensively investigate the heterogeneity of networks, as well as the diversity of service requests. On this basis, we formulate the heterogeneous network resource management (HNRM) problem as a multiple-objective problem, which aims at minimizing both the service delay and the network access cost simultaneously. Then, we propose a packet-encoding based multi-objective algorithm (PEMA), which consists of two components: packet encoding for data broadcast and multiobjective algorithm for network interface selection. Specifically, for improving bandwidth efficiency, we develop a multiple-packet encoding (MPE) technique to serve more requests simultaneously. For network selection, we propose a multi-objective evolutionary mechanism to further minimize both the service delay and the network access cost via population evolution. Finally, we give a comprehensive performance evaluation to demonstrate the superiority of PEMA under a wide range of scenarios.

Heterogeneous network resources are expected to cooperate with each other to support temporal data services in vehicular networks. However, it is challenging to implement an efficient data scheduling strategy due to the following factors: first, there are different time constraints on services, which are imposed by the application requirements of both temporal data quality and transmission delay; second, the heterogeneity of wireless interfaces further complicates the transmission task assignment in dynamic vehicular environments. Therefore, this paper proposes an software-defined network-based architecture to enable unified management on heterogeneous network resources. Then, we formulate the cooperative temporal data dissemination (CTDD) problem by considering the property of temporal data, the heterogeneity of wireless interfaces, and the delay constraints on service requests. Further, we prove the NP-hardness of the CTDD by constructing a polynomial-time reduction from a well know NP-hard problem, classical knapsack problem. On this basis, we design a heuristic algorithm called priority-based task assignment (PTA), which synthesizes dynamic task assignment, broadcast efficiency, and service deadline into priority design. Accordingly, PTA is able to adaptively distribute broadcast tasks of each request among multiple interfaces, so as to improve overall system performance. Last but not least, we build the simulation model and implement the proposed algorithm. The comprehensive simulation results show the superiority of the proposed algorithm under a wide range of scenarios.

Heterogeneous network resources are expected to cooperate with each other to support data services in vehicular networks. However, individual wireless interface cannot complete services within short vehicular dwelling time. Further, the network heterogeneity further complicates the transmission task assignment among multiple wireless interfaces. To resolve such an issue, we propose a novel architecture, where a scheduler is able to manage heterogeneous network resources in a centralized way. Then, we formulate the heterogeneous wireless interface management (HWIM) problem by considering both the heterogeneities of wireless interfaces and the delay constraints of service requests. On this basis, we design a heuristic algorithm called Adaptive Task Assignment (ATA), which synthesizes mobility feature, broadcast efficiency and service deadline into priority design. Accordingly, ATA is able to adaptively distribute broadcast task of each request among multiple interfaces, so as to improve overall system performance. Last but not the least, we build the simulation model and implement the proposed algorithm. The comprehensive simulation results show the superiority of the proposed algorithm.

In order to dynamically assign the optical/electrical routing path and optimize the optical/electrical resource allocation, a software-defined hybrid optical/electrical data center architecture (SD-HDC for short) is proposed in this work. A control mechanism that consists of two parallel steps including the new request handle step and the hybrid resource optimize step is introduced in SD-HDC. The new request handle step is used to quickly process the new incoming request while the hybrid resource optimize step periodically optimizes the hybrid optical/electrical network resource. To validate the SD-HDC architecture, a hybrid optical/electical routing algorithm used for handling elephant/mice flows is also proposed in this paper. Experimental results show that the proposed SD- HDC architecture can effectively reduce the blocking probability and path provisioning latency.

This paper studies the problem of how to efficiently minimize network coding resource. A modified particle swarm optimization (PSO) algorithm is proposed to tackle the problem, with the concept of path-relinking (PR) integrated into the evolutionary framework. As an efficient local search heuristic that makes use of problem-specific domain knowledge, PR helps strike a better balance between global exploration and local exploitation for the evolutionary search. Simulation results demonstrate that the proposed algorithm overweighs a number of existing and commonly used evolutionary algorithms (EAs) in terms of the solution quality, convergence, and computational time.

Because of the number of candidate nodes in opportunistic routing is too large, this paper proposed an estimation method on the number of candidate nodes based on distance (DBNCE). This method sets the number of candidate nodes for each node which participates in forwarding data package according to the distance between current node and the destination, also combines the two factors: network density and the number of neighbor nodes in current node. Simulation results show that using DBNCE in opportunistic routing will reduce the number of candidate nodes effectively while guarantee the rate of data transmission, and improve the performance of the network.

This paper presents a modified ant colony optimization (ACO) approach for the network coding resource minimization problem. It is featured with several attractive mechanisms specially devised for solving the concerned problem: 1) a multidimensional pheromone maintenance mechanism is put forward to address the issue of pheromone overlapping; 2) problem-specific heuristic information is employed to enhance the capability of heuristic search (neighboring area search); 3) a tabu-table-based path construction method is devised to facilitate the construction of feasible (link-disjoint) paths from the source to each receiver; 4) a local pheromone updating rule is developed to guide ants to construct appropriate promising paths; and 5) a solution reconstruction method is presented, with the aim of avoiding prematurity and improving the global search efficiency of proposed algorithm. Due to the way it works, the ACO can well exploit the global and local information of routing-related problems during the solution construction phase. The simulation results on benchmark instances demonstrate that with the integrated five extended mechanisms, our algorithm outperforms a number of existing algorithms with respect to the best solutions obtained and the computational time.