While vision-language foundation models (VLMs) achieve remarkable performance when fine-tuned on downstream in-distribution (ID) data, this process compromises their generalization ability on out-of-distribution (OOD) data that deviate from the downstream tasks due to overfitting. To address this, we propose ProLoG, a new adaptation method that effectively fine-tunes VLMs on downstream tasks while achieving high OOD performance. Specifically, we design a unique integration of prompt tuning and LoRA, offering a robust hybrid platform to improve performance. During training, we propose an augmentation-based regularization loss that enhances the generalization of our hybrid network by using augmented image features aligned with LLM-generated texts containing key attributes of each class. By leveraging our hybrid design, we also introduce an adaptive inference strategy that flexibly applies trained prompts and LoRA based on a task similarity score to effectively handle both ID and OOD data. Experimental results demonstrate that our proposed method outperforms existing works on various datasets, confirming its advantages.

Much of federated learning (FL) focuses on settings where local dataset statistics remain the same between training and testing. However, this assumption often does not hold in practice due to distribution shifts, motivating the development of domain generalization (DG) approaches that leverage source domain data to train models capable of generalizing to unseen target domains. In this paper, we are motivated by two major gaps in existing work on FL and DG: (1) the lack of formal mathematical analysis of DG objectives; and (2) DG research in FL being limited to the star-topology architecture. We develop Decentralized Federated Domain Generalization with Style Sharing ($\textit{StyleDDG}$), a decentralized DG algorithm which allows devices in a peer-to-peer network to achieve DG based on sharing style information inferred from their datasets. Additionally, we provide the first systematic approach to analyzing style-based DG training in decentralized networks. We cast existing centralized DG algorithms within our framework, and employ their formalisms to model $\textit{StyleDDG}$. We then obtain analytical conditions under which convergence of $\textit{StyleDDG}$ can be guaranteed. Through experiments on popular DG datasets, we demonstrate that $\textit{StyleDDG}$ can obtain significant improvements in accuracy across target domains with minimal communication overhead compared to baseline decentralized gradient methods.

Multimodal federated learning (MFL) aims to enrich model training in FL settings where clients are collecting measurements across multiple modalities. However, key challenges to MFL remain unaddressed, particularly in heterogeneous network settings where: (i) the set of modalities collected by each client is diverse, and (ii) communication limitations prevent clients from uploading all their locally trained modality encoders to the server. In this paper, we propose Multimodal Federated learning with joint Modality and Client selection (MFedMC), a communication-efficient MFL framework that tackles these challenges through a decoupled architecture and selective uploading. Unlike traditional holistic fusion approaches, MFedMC separates modality encoders and fusion modules: modality encoders are aggregated at the server for generalization across diverse client distributions, while fusion modules remain local to each client for personalized adaptation to individual modality configurations and data characteristics. Building on this decoupled design, our joint selection algorithm incorporates two main components: (a) A modality selection methodology for each client, which weighs (i) the impact of the modality, gauged by Shapley value analysis, (ii) the modality encoder size as a gauge of communication overhead, and (iii) the frequency of modality encoder updates, denoted recency, to enhance generalizability. (b) A client selection strategy for the server based on the local loss of modality encoders at each client. Experiments on five real-world datasets demonstrate that MFedMC achieves comparable accuracy to several baselines while reducing communication overhead by over 20×. A demo video and our code are available athttps://liangqiy.com/mfedmc/.

While federated learning (FL) eliminates the transmission of raw data over a network, it is still vulnerable to privacy breaches from the communicated model parameters. Differential privacy (DP) is often employed to address such issues. However, the impact of DP on FL in multi-tier networks – where hierarchical aggregations couple noise injection decisions at different tiers, and trust models are heterogeneous across subnetworks–is not well understood. To fill this gap, we develop Multi-Tier Federated Learning with Multi-Tier Differential Privacy (M2FDP), a DP-enhanced FL methodology for jointly optimizing privacy and performance over such networks. One of the key principles ofM2FDPis to adapt DP noise injection across the established edge/fog computing hierarchy (e.g., edge devices, intermediate nodes, and other tiers up to cloud servers) according to the trust models in different subnetworks. We conduct a comprehensive analysis of the convergence behavior ofM2FDPunder non-convex problem settings, revealing conditions on parameter tuning under which the training process converges sublinearly to a finite stationarity gap that depends on the network hierarchy, trust model, and target privacy level. We show how these relationships can be employed to develop an adaptive control algorithm forM2FDPthat tunes properties of local model training to minimize energy, latency, and the stationarity gap while meeting desired convergence and privacy criterion. Subsequent numerical evaluations demonstrate thatM2FDPobtains substantial improvements in these metrics over baselines for different privacy budgets and system configurations.

In multi-cell wireless networks, providing intelligent services via federated learning (FL) becomes more challenging due to multi-level distribution shifts across clients and regions, as well as additional communication delays among edge and cloud servers. To address these issues, we propose SplitOMC, a split learning framework that integrates overlapping-area clients and a multi-exit neural architecture to jointly handle (i) client-preferred, (ii) out-of-preference, and (iii) out-of-region tasks. By strategically leveraging clients in overlapping regions, SplitOMC accelerates training without excessive backhaul communication, while maintaining both personalization and generalization. We theoretically analyze the convergence behavior of the proposed algorithm, ensuring performance stability under heterogeneous data and communication conditions. Extensive experiments on MNIST, CIFAR-10/100, and a real-world Jetson Nano testbed demonstrate that SplitOMC consistently achieves faster training and inference with improved accuracy compared to state-of-the-art methods. In particular, the framework shows robustness in resource-constrained and unstable network environments, highlighting its practical value for next-generation wireless intelligent services.

A few recent studies have shown the benefits of using centrally pre-trained models to initialize federated learning (FL). However, existing methods do not generalize well when faced with an arbitrary set of downstream FL tasks. Specifically, they often (i) achieve limited accuracy, especially with unseen downstream labels, and (ii) result in significant accuracy variance, failing to provide a balanced performance across clients. To address these challenges, we propose CoPreFL, a collaborative/distributed pre-training approach that robustly initializes for downstream FL tasks. CoPreFL leverages model-agnostic meta-learning (MAML) that tailors the global model to mimic heterogeneous and unseen FL scenarios, resulting in a pre-trained model that is rapidly adaptable to any FL task. Our MAML procedure integrates performance variance into the meta-objective function, balancing performance across clients rather than solely optimizing for accuracy. Extensive experiments show that CoPreFL significantly enhances average accuracy and reduces variance in arbitrary downstream FL tasks with unseen/seen labels, outperforming various pre-training baselines. Additionally, CoPreFL proves compatible with different well-known FL algorithms used in downstream tasks, boosting performance in each case.

In response to the increasing demand for tackling out-of-domain (OOD) scenarios, test-time adaptation (TTA) has garnered significant research attention in recent years. To adapt a source pre-trained model to target samples without getting access to their labels, existing approaches have typically employed entropy minimization (EM) loss as a primary objective function. In this paper, we propose an adaptive energy alignment (AEA) solution that achieves fast online TTA. We start from the re-interpretation of the EM loss by decomposing it into two energy-based terms with conflicting roles, showing that the EM loss can potentially hinder the assertive model adaptation. Our AEA addresses this challenge by strategically reducing the energy gap between the source and target domains during TTA, aiming to effectively align the target domain with the source domains and thus to accelerate adaptation. We specifically propose two novel strategies, each contributing a necessary component for TTA: (i) aligning the energy level of each target sample with the energy zone of the source domain that the pre-trained model is already familiar with, and (ii) precisely guiding the direction of the energy alignment by matching the class-wise correlations between the source and target domains. Our approach demonstrates its effectiveness on various domain shift datasets including CIFAR10-C, CIFAR100-C, and TinyImageNet-C.

Over the past several years, various federated learning (FL) methodologies have been developed to improve model accuracy, a primary performance metric in machine learning. However, to utilize FL in practical decision-making scenarios, beyond considering accuracy, the trained model must also have a reliable confidence in each of its predictions, an aspect that has been largely overlooked in existing FL research. Motivated by this gap, we propose Non-Uniform Calibration for Federated Learning (NUCFL), a generic framework that integrates FL with the concept of model calibration. The inherent data heterogeneity in FL environments makes model calibration particularly difficult, as it must ensure reliability across diverse data distributions and client conditions. Our NUCFL addresses this challenge by dynamically adjusting the model calibration objectives based on statistical relationships between each client's local model and the global model in FL. In particular, NUCFL assesses the similarity between local and global model relationships, and controls the penalty term for the calibration loss during client-side local training. By doing so, NUCFL effectively aligns calibration needs for the global model in heterogeneous FL settings while not sacrificing accuracy. Extensive experiments show that NUCFL offers flexibility and effectiveness across various FL algorithms, enhancing accuracy as well as model calibration.

To improve the efficiency of reinforcement learning (RL), we propose a novel asynchronous federated reinforcement learning (FedRL) framework termed AFedPG, which constructs a global model through collaboration among $N$ agents using policy gradient (PG) updates. To address the challenge of lagged policies in asynchronous settings, we design a delay-adaptive lookahead technique *specifically for FedRL* that can effectively handle heterogeneous arrival times of policy gradients. We analyze the theoretical global convergence bound of AFedPG, and characterize the advantage of the proposed algorithm in terms of both the sample complexity and time complexity. Specifically, our AFedPG method achieves $\mathcal{O}(\frac{{\epsilon}^{-2.5}}{N})$ sample complexity for global convergence at each agent on average. Compared to the single agent setting with $\mathcal{O}(\epsilon^{-2.5})$ sample complexity, it enjoys a linear speedup with respect to the number of agents. Moreover, compared to synchronous FedPG, AFedPG improves the time complexity from $\mathcal{O}(\frac{t_{\max}}{N})$ to $\mathcal{O}({\sum_{i=1}^{N} \frac{1}{t_{i}}})^{-1}$, where $t_{i}$ denotes the time consumption in each iteration at agent $i$, and $t_{\max}$ is the largest one. The latter complexity $\mathcal{O}({\sum_{i=1}^{N} \frac{1}{t_{i}}})^{-1}$ is always smaller than the former one, and this improvement becomes significant in large-scale federated settings with heterogeneous computing powers ($t_{\max}\gg t_{\min}$). Finally, we empirically verify the improved performance of AFedPG in four widely used MuJoCo environments with varying numbers of agents. We also demonstrate the advantages of AFedPG in various computing heterogeneity scenarios.

Despite recent advancements in federated learning (FL), the integration of generative models into FL has been limited due to challenges such as high communication costs and unstable training in heterogeneous data environments. To address these issues, we propose PRISM, a FL framework tailored for generative models that ensures (i) stable performance in heterogeneous data distributions and (ii) resource efficiency in terms of communication cost and final model size. The key of our method is to search for an optimal stochastic binary mask for a random network rather than updating the model weights, identifying a sparse subnetwork with high generative performance; i.e., a ``strong lottery ticket''. By communicating binary masks in a stochastic manner, PRISM minimizes communication overhead. This approach, combined with the utilization of maximum mean discrepancy (MMD) loss and a mask-aware dynamic moving average aggregation method (MADA) on the server side, facilitates stable and strong generative capabilities by mitigating local divergence in FL scenarios. Moreover, thanks to its sparsifying characteristic, PRISM yields a lightweight model without extra pruning or quantization, making it ideal for environments such as edge devices. Experiments on MNIST, FMNIST, CelebA, and CIFAR10 demonstrate that PRISM outperforms existing methods, while maintaining privacy with minimal communication costs. PRISM is the first to successfully generate images under challenging non-IID and privacy-preserving FL environments on complex datasets, where previous methods have struggled.

Decentralized federated learning (DFL) captures FL settings where both (i) model updates and (ii) model aggregations are exclusively carried out by the clients without a central server. Existing DFL works have mostly focused on settings where clients conduct a fixed number of local updates between local model exchanges, overlooking heterogeneity and dynamics in communication and computation capabilities. In this work, we propose Decentralized Sporadic Federated Learning ($\texttt{DSpodFL}$), a DFL methodology built on a generalized notion of *sporadicity* in both local gradient and aggregation processes. $\texttt{DSpodFL}$ subsumes many existing decentralized optimization methods under a unified algorithmic framework by modeling the per-iteration (i) occurrence of gradient descent at each client and (ii) exchange of models between client pairs as arbitrary indicator random variables, thus capturing *heterogeneous and time-varying* computation/communication scenarios. We analytically characterize the convergence behavior of $\texttt{DSpodFL}$ for both convex and non-convex models and for both constant and diminishing learning rates, under mild assumptions on the communication graph connectivity, data heterogeneity across clients, and gradient noises. We show how our bounds recover existing results from decentralized gradient descent as special cases. Experiments demonstrate that $\texttt{DSpodFL}$ consistently achieves improved training speeds compared with baselines under various system settings.

Compared to traditional machine learning models, recent large language models (LLMs) can exhibit multi-task-solving capabilities through multiple dialogues and multi-modal data sources. These unique characteristics of LLMs, together with their large model size, make their deployment more challenging. Specifically, (i) deploying LLMs on local devices faces computational, memory, and energy resource issues, while (ii) deploying them in the cloud cannot guarantee real-time service and incurs communication/usage costs. In this paper, we design TMO, a local-cloud LLM inference system with Three-M Offloading: Multi-modal, Multi-task, and Multi-dialogue. TMO incorporates (i) a lightweight local LLM that can process simple tasks at high speed and (ii) a large-scale cloud LLM that can handle multi-modal data sources. We develop a resource-constrained reinforcement learning (RCRL) strategy for TMO that optimizes the inference location (i.e., local vs. cloud) and multi-modal data sources to use for each task/dialogue, aiming to maximize the long-term reward (response quality, latency, and usage cost) while adhering to resource constraints. We also contribute M4A1, a new dataset we curated that contains reward and cost metrics across multiple modality, task, dialogue, and LLM configurations, enabling evaluation of offloading decisions. We demonstrate the effectiveness of TMO compared to several exploration-decision and LLM-as-Agent baselines, showing significant improvements in latency, cost, and response quality.

Hierarchical federated learning (HFL) has demonstrated promising scalability advantages over the traditional “star-topology” architecture-based federated learning (FL). However, HFL still imposes significant computation, communication, and storage burdens on the edge, especially when training a large-scale model over resource-constrained wireless devices. In this paper, we propose hierarchical independent submodel training (HIST), a new FL methodology that aims to address these issues in hierarchical cloud-edge-client networks. The key idea behindHISTis to divide the global model into disjoint partitions (or submodels) per round so that each group of clients (i.e., cells) is responsible for training only one partition of the model. We characterize the convergence behavior ofHISTunder mild assumptions, showing the impacts of several key attributes (e.g., submodel sizes, number of cells, edge and global aggregation frequencies) on the rate and stationarity gap. Building upon the theoretical results, we propose a submodel partitioning strategy to minimize the training latency depending on network resource availability and a target learning performance guarantee. We then demonstrate howHISTcan be augmented with over-the-air computation (AirComp) to further enhance the efficiency of the model aggregation over the edge cells. Through numerical evaluations, we verify thatHISTis able to save training time and communication costs by wide margins while achieving comparable accuracy as conventional HFL. Moreover, our experiments demonstrate that AirComp-assistedHISTprovides further improvements in training latency.

Research interests in the robustness of deep neural networks against domain shifts have been rapidly increasing in recent years. Most existing works, however, focus on improving the accuracy of the model, not the calibration performance which is another important requirement for trustworthy AI systems. Temperature scaling (TS), an accuracy-preserving post-hoc calibration method, has been proven to be effective in in-domain settings, but not in out-of-domain (OOD) due to the difficulty in obtaining a validation set for the unseen domain beforehand. In this paper, we propose consistency-guided temperature scaling (CTS), a new temperature scaling strategy that can significantly enhance the OOD calibration performance by providing mutual supervision among data samples in the source domains. Motivated by our observation that over-confidence stemming from inconsistent sample predictions is the main obstacle to OOD calibration, we propose to guide the scaling process by taking consistencies into account in terms of two different aspects - style and content - which are the key components that can well-represent data samples in multi-domain settings. Experimental results demonstrate that our proposed strategy outperforms existing works, achieving superior OOD calibration performance on various datasets. This can be accomplished by employing only the source domains without compromising accuracy, making our scheme directly applicable to various trustworthy AI systems.

Hierarchical federated learning (HFL) has demon-strated promising scalability advantages over the traditional “star-topology” architecture-based federated learning (FL). How-ever, HFL still imposes significant computation, communication, and storage burdens on the edge, especially when training a large-scale model over resource-constrained Internet of Things (IoT) devices. In this paper, we propose hierarchical independent submodel training (HIST), a new FL methodology that aims to address these issues in hierarchical settings. The key idea behind HIST is a hierarchical version of model partitioning, where we partition the global model into disjoint submodels in each round, and distribute them across different cells, so that each cell is responsible for training only one partition of the full model. This enables each client to save computation/storage costs while alleviating the communication loads throughout the hierarchy. We characterize the convergence behavior of HIST for non-convex loss functions under mild assumptions, showing the impact of several attributes (e.g., number of cells, local and global aggregation frequency) on the performance-efficiency tradeoff. Finally, through numerical experiments, we verify that HIST is able to save communication costs by a wide margin while achieving the same target testing accuracy.

While network coverage maps continue to expand, many devices located in remote areas remain unconnected to terrestrial communication infrastructures, preventing them from getting access to the associated data-driven services. In this paper, we propose a cooperative ground-to-satellite federated learning (FL) methodology to facilitate machine learning service management over remote regions. Our methodology orchestrates satellite constellations to provide the following key functions during FL: (i) processing data offloaded from ground devices, (ii) aggregating models within device clusters, and (iii) relaying models/data to other satellites via inter-satellite links (ISLs). Due to the limited coverage time of each satellite over a particular remote area, we facilitate satellite transmission of trained models and acquired data to neighboring satellites via ISL, so that the incoming satellite can continue FL for the region. We also develop a training latency minimizer which optimizes over the amount of data to be offloaded from ground devices to satellites. Through experiments on benchmark datasets, we show that our scheme can significantly speed up the convergence of FL compared with terrestrial-only and other satellite baseline approaches.

Multimodal federated learning (FL) aims to enrich model training in FL settings where devices are collecting measurements across multiple modalities (e.g., sensors measuring pressure, motion, and other types of data). However, key challenges to multimodal FL remain unaddressed, particularly in heterogeneous network settings: (i) the set of modalities collected by each device will be diverse, and (ii) communication limitations prevent devices from uploading all their locally trained modality models to the server. In this paper, we propose Federated Multimodal Fusion learning with Selective modality communication (FedMFS), a new multimodal fusion FL methodology that can tackle the above mentioned challenges. The key idea is the introduction of a modality selection criterion for each device, which weighs (i) the impact of the modality, gauged by Shapley value analysis, against (ii) the modality model size as a gauge for communication overhead. This enables FedMFS to flexibly balance performance against communication costs, depending on resource constraints and application requirements. Experiments on the real-world ActionSense dataset demonstrate the ability of FedMFS to achieve comparable accuracy to several baselines while reducing the communication overhead by over 4x.

Handling the substantial communication burden in federated learning (FL) still remains a significant challenge. Although recent studies have attempted to compress the local gradients to address this issue, they typically perform compression only within the original parameter space, which may potentially limit the fundamental compression rate of the gradient. In this paper, instead of restricting our scope to a fixed traditional space, we consider an alternative space that provides an improved compressibility of the gradient. To this end, we utilize the structures of input activation and output gradient in designing our mapping function to a new space, which enables *lossless gradient sparsification*, i.e., mapping the gradient to our new space induces a greater number of *near-zero* elements without any loss of information. In light of this attribute, employing sparsification-based compressors in our new space allows for more aggressive compression with minimal information loss than the baselines. More surprisingly, our model even reaches higher accuracies than the full gradient uploading strategy in some cases, an extra benefit for utilizing the new space. We also theoretically confirm that our approach does not alter the existing, best known convergence rate of FL thanks to the orthogonal transformation properties of our mapping.

While traditional federated learning (FL) typically focuses on a star topology where clients are directly connected to a central server, real-world distributed systems often exhibit hierarchical architectures. Hierarchical FL (HFL) has emerged as a promising solution to bridge this gap, leveraging aggregation points at multiple levels of the system. However, existing algorithms for HFL encounter challenges in dealing with multi-timescale model drift, i.e., model drift occurring across hierarchical levels of data heterogeneity. In this paper, we propose a multi-timescale gradient correction (MTGC) methodology to resolve this issue. Our key idea is to introduce distinct control variables to (i) correct the client gradient towards the group gradient, i.e., to reduce client model drift caused by local updates based on individual datasets, and (ii) correct the group gradient towards the global gradient, i.e., to reduce group model drift caused by FL over clients within the group. We analytically characterize the convergence behavior of MTGC under general non-convex settings, overcoming challenges associated with couplings between correction terms. We show that our convergence bound is immune to the extent of data heterogeneity, confirming the stability of the proposed algorithm against multi-level non-i.i.d. data. Through extensive experiments on various datasets and models, we validate the effectiveness of MTGC in diverse HFL settings. The code for this project is available at https://github.com/wenzhifang/MTGC.

Devices located in remote regions often lack coverage from well-developed terrestrial communication infrastructure. This not only prevents them from experiencing high quality communication services but also hinders the delivery of machine learning services in remote regions. In this paper, we propose a new federated learning (FL) methodology tailored to space-air-ground integrated networks (SAGINs) to tackle this issue. Our approach strategically leverages the nodes within space and air layers as both 1) edge computing units and 2) model aggregators during the FL process, addressing the challenges that arise from the limited computation powers of ground devices and the absence of terrestrial base stations in the target region. The key idea behind our methodology is the adaptive data offloading and handover procedures that incorporate various network dynamics in SAGINs, including the mobility, heterogeneous computation powers, and inconsistent coverage times of incoming satellites. We analyze the latency of our scheme and develop an adaptive data offloading optimizer, and also characterize the theoretical convergence bound of our proposed algorithm. Experimental results confirm the advantage of our SAGIN-assisted FL methodology in terms of training time and test accuracy compared with various baselines.

While network coverage maps continue to expand, many devices located in remote areas remain unconnected to terrestrial communication infrastructures, preventing them from getting access to the associated data-driven services. In this paper, we propose a ground-to-satellite cooperative federated learning (FL) methodology to facilitate machine learning service management over remote regions. Our methodology orchestrates satellite constellations to provide the following key functions during FL: (i) processing data offloaded from ground devices, (ii) aggregating models within device clusters, and (iii) relaying models/data to other satellites via inter-satellite links (ISLs). Due to the limited coverage time of each satellite over a particular remote area, we facilitate satellite transmission of trained models and acquired data to neighboring satellites via ISL, so that the incoming satellite can continue conducting FL for the region. We theoretically analyze the convergence behavior of our algorithm, and develop a training latency minimizer which optimizes over satellite-specific network resources, including the amount of data to be offloaded from ground devices to satellites and satellites’ computation speeds. Through experiments on three datasets, we show that our methodology can significantly speed up the convergence of FL compared with terrestrial-only and other satellite baseline approaches.

The demand for intelligent services at the network edge has introduced several research challenges. One is the need for a machine learning architecture that achieves personalization (to individual clients) and generalization (to unseen data) properties concurrently across different applications. Another is the need for an inference strategy that can satisfy network resource and latency constraints during testing-time. Existing techniques in federated learning have encountered a steep trade-off between personalization and generalization, and have not explicitly considered the resource requirements during the inference-stage. In this paper, we propose SplitGP, a joint edge-AI training and inference strategy that simultaneously captures generalization/personalization for efficient inference across resource-constrained clients. The training process of SplitGP is based on federated split learning, with the key idea of optimizing the client-side model to have personalization capability tailored to its main task, while training the server-side model to have generalization capability for handling out-of-distribution tasks. During testing-time, each client selectively offloads inference tasks to the server based on the uncertainty threshold tunable based on network resource availability. Through formal convergence analysis and inference time analysis, we provide guidelines on the selection of key meta-parameters in SplitGP. Experimental results confirm the advantage of SplitGP over existing baselines.

As the size of a model increases, making predictions using deep neural networks (DNNs) is becoming more computationally expensive. Multi-exit neural network is one promising solution that can flexibly make anytime predictions via early exits, depending on the current test-time budget which may vary over time in practice (e.g., self-driving cars with dynamically changing speeds). However, the prediction performance at the earlier exits is generally much lower than the final exit, which becomes a critical issue in low-latency applications having a tight test-time budget. Compared to the previous works where each block is optimized to minimize the losses of all exits simultaneously, in this work, we propose a new method for training multi-exit neural networks by strategically imposing different objectives on individual blocks. The proposed idea based on grouping and overlapping strategies improves the prediction performance at the earlier exits while not degrading the performance of later ones, making our scheme to be more suitable for low-latency applications. Extensive experimental results on both image classification and semantic segmentation confirm the advantage of our approach. The proposed idea does not require any modifications in the model architecture and can be easily combined with existing strategies aiming to improve the performance of multi-exit neural networks.

In domain generalization (DG), the target domain is unknown when the model is being trained, and the trained model should successfully work on an arbitrary (and possibly unseen) target domain during inference. This is a difficult problem, and despite active studies in recent years, it remains a great challenge. In this paper, we take a simple yet effective approach to tackle this issue. We propose test-time style shifting, which shifts the style of the test sample (that has a large style gap with the source domains) to the nearest source domain that the model is already familiar with, before making the prediction. This strategy enables the model to handle any target domains with arbitrary style statistics, without additional model update at test-time. Additionally, we propose style balancing, which provides a great platform for maximizing the advantage of test-time style shifting by handling the DG-specific imbalance issues. The proposed ideas are easy to implement and successfully work in conjunction with various other DG schemes. Experimental results on different datasets show the effectiveness of our methods.

A fundamental challenge to providing edge-AI services is the need for a machine learning (ML) model that achieves personalization (i.e., to individual clients) and generalization (i.e., to unseen data) properties concurrently. Existing techniques in federated learning (FL) have encountered a steep tradeoff between these objectives and impose large computational requirements on edge devices during training and inference. In this paper, we propose SplitGP, a new split learning solution that can simultaneously capture generalization and personalization capabilities for efficient inference across resource-constrained clients (e.g., mobile/IoT devices). Our key idea is to split the full ML model into client-side and server-side components, and impose different roles to them: the client-side model is trained to have strong personalization capability optimized to each client’s main task, while the server-side model is trained to have strong generalization capability for handling all clients’ out-of-distribution tasks. We analytically characterize the convergence behavior of SplitGP, revealing that all client models approach stationary points asymptotically. Further, we analyze the inference time in SplitGP and provide bounds for determining model split ratios. Experimental results show that SplitGP outperforms existing baselines by wide margins in inference time and test accuracy for varying amounts of out-of-distribution samples.

While multi-exit neural networks are regarded as a promising solution for making efficient inference via early exits, combating adversarial attacks remains a challenging problem. In multi-exit networks, due to the high dependency among different submodels, an adversarial example targeting a specific exit not only degrades the performance of the target exit but also reduces the performance of all other exits concurrently. This makes multi-exit networks highly vulnerable to simple adversarial attacks. In this paper, we propose NEO-KD, a knowledge-distillation-based adversarial training strategy that tackles this fundamental challenge based on two key contributions. NEO-KD first resorts to neighbor knowledge distillation to guide the output of the adversarial examples to tend to the ensemble outputs of neighbor exits of clean data. NEO-KD also employs exit-wise orthogonal knowledge distillation for reducing adversarial transferability across different submodels. The result is a significantly improved robustness against adversarial attacks. Experimental results on various datasets/models show that our method achieves the best adversarial accuracy with reduced computation budgets, compared to the baselines relying on existing adversarial training or knowledge distillation techniques for multi-exit networks.

Traditional federated learning (FL) algorithms operate under the assumption that the data distributions at training (source domains) and testing (target domain) are the same. The fact that domain shifts often occur in practice necessitates equipping FL methods with a domain generalization (DG) capability. However, existing DG algorithms face fundamental challenges in FL setups due to the lack of samples/domains in each client’s local dataset. In this paper, we propose StableFDG, a style and attention based learning strategy for accomplishing federated domain generalization, introducing two key contributions. The first is style-based learning, which enables each client to explore novel styles beyond the original source domains in its local dataset, improving domain diversity based on the proposed style sharing, shifting, and exploration strategies. Our second contribution is an attention-based feature highlighter, which captures the similarities between the features of data samples in the same class, and emphasizes the important/common characteristics to better learn the domain-invariant characteristics of each class in data-poor FL scenarios. Experimental results show that StableFDG outperforms existing baselines on various DG benchmark datasets, demonstrating its efficacy.

Recent proliferation of mobile devices and edge servers (e.g., small base stations) strongly motivates distributed learning at the wireless edge. In this paper, we propose a fast and secure distributed learning framework that utilizes computing resources at edge servers as well as distributed computing devices in tiered wireless edge networks. A fundamental lower bound is derived on the computational load that perfectly tolerates Byzantine attacks at both tiers. TiBroco, a hierarchical coding framework achieving this theoretically minimum computational load is proposed, which guarantees secure distributed learning by combating Byzantines. A fast distributed learning is possible by precisely allocating loads to the computing devices and edge servers, and also utilizing the broadcast nature of wireless devices. Extensive experimental results on Amazon EC2 indicate that our TiBroco allows significantly faster distributed learning than existing methods while guaranteeing full tolerance against Byzantine attacks at both tiers.

While federated learning (FL) allows efficient model training with local data at edge devices, among major issues still to be resolved are: slow devices known as stragglers and malicious attacks launched by adversaries. While the presence of both of these issues raises serious concerns in practical FL systems, no known schemes or combinations of schemes effectively address them at the same time. We propose Sageflow, staleness-aware grouping with entropy-based filtering and loss-weighted averaging, to handle both stragglers and adversaries simultaneously. Model grouping and weighting according to staleness (arrival delay) provides robustness against stragglers, while entropy-based filtering and loss-weighted averaging, working in a highly complementary fashion at each grouping stage, counter a wide range of adversary attacks. A theoretical bound is established to provide key insights into the convergence behavior of Sageflow. Extensive experimental results show that Sageflow outperforms various existing methods aiming to handle stragglers/adversaries.

In federated learning (FL), a number of distributed clients targeting the same task collaborate to train a single global model without sharing their data. The learning process typically starts from a randomly initialized or some pretrained model. In this paper, we aim at designing an initial model based on which an arbitrary group of clients can obtain a global model for its own purpose, within only a few rounds of FL. The key challenge here is that the downstream tasks for which the pretrained model will be used are generally unknown when the initial model is prepared. Our idea is to take a meta-learning approach to construct the initial model so that any group with a possibly unseen task can obtain a high-accuracy global model within only R rounds of FL. Our meta-learning itself could be done via federated learning among willing participants and is based on an episodic arrangement to mimic the R rounds of FL followed by inference in each episode. Extensive experimental results show that our method generalizes well for arbitrary groups of clients and provides large performance improvements given the same overall communication/computation resources, compared to other baselines relying on known pretraining methods.

We consider federated learning (FL) with multiple wireless edge servers having their own local coverage. We focus on speeding up training in this increasingly practical setup. Our key idea is to utilize the clients located in the overlapping coverage areas among adjacent edge servers (ESs); in the model-downloading stage, the clients in the overlapping areas receive multiple models from different ESs, take the average of the received models, and then update the averaged model with their local data. These clients send their updated model to multiple ESs by broadcasting, which acts as bridges for sharing the trained models between servers. Even when some ESs are given biased datasets within their coverage regions, their training processes can be assisted by adjacent servers through the clients in their overlapping regions. As a result, the proposed scheme does not require costly communications with the central cloud server (located at the higher tier of edge servers) for model synchronization, significantly reducing the overall training time compared to the conventional cloud-based FL systems. Extensive experimental results show remarkable performance gains of our scheme compared to existing methods. Our design targets latency-sensitive applications where edge-based FL is essential, e.g., when a number of connected cars/drones must cooperate (via FL) to quickly adapt to dynamically changing environments.

Wireless caching networks have been extensively researched as a promising technique for supporting the massive data traffic of multimedia services. Many of the existing studies on real-data traffic have shown that users of a multimedia service consecutively request multiple contents and this sequence is strongly dependent on the related list of the first content and/or the top referrer in the category. This paper thus introduces the notion of “temporary preference”, characterizing the behavior of users who are highly likely to request the next content from a certain target category (i.e., related content list). Based on this observation, this paper proposes both probabilistic caching and dynamic delivery policies for categorized contents and consecutive user demands. The proposed caching scheme maximizes the minimum of the cache hit rates for all users. In the delivery phase, a dynamic helper association policy for receiving multiple contents in a row is designed to reduce the delivery latency. By comparing with the content placement optimized for one-shot requests, numerical results verify the effects of categorized contents and consecutive user demands on the proposed caching and delivery policies.

Distributed learning plays a key role in reducing the training time of modern deep neural networks with massive datasets. In this article, we consider a distributed learning problem where gradient computation is carried out over a number of computing devices at the wireless edge. We propose hierarchical broadcast coding, a provable coding-theoretic framework to speed up distributed learning at the wireless edge. Our contributions are threefold. First, motivated by the hierarchical nature of real-world edge computing systems, we propose a layered code which mitigates the effects of not only packet losses at the wireless computing nodes but also straggling access points (APs) or small base stations. Second, by strategically allocating data partitions to nodes in the overlapping areas between cells, our technique achieves the fundamental lower bound on computational load to combat stragglers. Finally, we take advantage of the broadcast nature of wireless networks by which wireless devices in the overlapping cell coverage broadcast to more than one AP. This further reduces the overall training time in the presence of straggling APs. Experimental results on Amazon EC2 confirm the advantage of the proposed methods in speeding up learning. Our design targets any gradient descent based learning algorithms, including linear/logistic regressions and deep learning.

In wireless distributed computing systems, mobile devices that are connected wirelessly to the Fog (e.g., small base stations) collaboratively solve a given computational task. Unfortunately, wireless distributed computing systems suffer from packet losses due to severe channel fading. Moreover, a wireless device can drop out of the system when leaving the coverage of a master node in the Fog layer. We model this unreliability between a device and a master node as a packet erasure channel. When a packet fails to be detected at the receiver, the corresponding packet is retransmitted, which would significantly increase the overall run-time to finish the task. We take a coding-theoretic approach to tackle this straggler-like problem in wireless distributed computing. We first investigate the expected latency using an$(n, k)$maximum-distance separable (MDS) code. We obtain the lower and upper bounds on the latency in closed forms and provide guidelines to design MDS codes depending on thechannelcondition characterized by packet erasure probability. Then, we introduce another important performance metric calledminimum latency, and also provide guidelines on designing optimal codes. Based on optimal codes, we obtain the performance curves of achievable minimum latency and achievable workload as functions of packet erasure probability.

Current distributed learning systems suffer from serious performance degradation under Byzantine attacks. This paper proposes Election Coding, a coding-theoretic framework to guarantee Byzantine-robustness for distributed learning algorithms based on signed stochastic gradient descent (SignSGD) that minimizes the worker-master communication load. The suggested framework explores new information-theoretic limits of finding the majority opinion when some workers could be attacked by adversary, and paves the road to implement robust and communication-efficient distributed learning algorithms. Under this framework, we construct two types of codes, random Bernoulli codes and deterministic algebraic codes, that tolerate Byzantine attacks with a controlled amount of computational redundancy and guarantee convergence in general non-convex scenarios. For the Bernoulli codes, we provide an upper bound on the error probability in estimating the signs of the true gradients, which gives useful insights into code design for Byzantine tolerance. The proposed deterministic codes are proven to perfectly tolerate arbitrary Byzantine attacks. Experiments on real datasets confirm that the suggested codes provide substantial improvement in Byzantine tolerance of distributed learning systems employing SignSGD.

In wireless caching networks, a user generally has a concrete purpose of consuming contents in a certain preferred category, and requests multiple contents in sequence. While most existing research on wireless caching and delivery has focused only on one-shot requests, the popularity distribution of contents requested consecutively is definitely different from the one-shot request and has been not considered. Also, especially from the perspective of the service provider, it is advantageous for users to consume as many contents as possible. Thus, this paper proposes two cache allocation policies for categorized contents and consecutive user demands, which maximize 1) the cache hit rate and 2) the number of consecutive content consumption, respectively. Numerical results show how categorized contents and consecutive content requests have impacts on the cache allocation.

In wireless caching networks, each user generally consumes more than one content in a row, and the number of consecutive demands could vary for different users. In addition, popular contents are usually classified into several categories. In this case for consecutive user demands, the content popularity model largely depends on the previously consumed contents, i.e., contents that belong to the same category as the previously consumed content would be highly popular. Based on this observation, this paper proposes an optimal probabilistic caching policy for consecutive user demands in categorized contents. The proposed caching scheme maximizes the minimum of the success probabilities for content delivery of all users when individual users request different numbers of contents in a row. Comparing with the content placement optimized for one-shot request, intensive numerical results verify the impacts of categorized contents and consecutive user demands on the caching policy.

We suggest a general framework for network-coded Practical Byzantine Fault Tolerant (PBFT) consensus for enabling agreement among distributed nodes under Byzantine attacks. The suggested protocol generalizes existing replication and sharding schemes which are frequently used for consensus in current blockchain systems. Using the proposed algorithm, it is possible to reach a consensus when the available bandwidth is considerably smaller on individual links compared to that required for conventional schemes. It is shown that there exists an upper bound on the number of nodes that can participate in the protocol, given a maximum bandwidth constraint across all pairwise links. Furthermore, the protocol that achieves the upper bound is provided by using a set of constant weight codes.

Coded computation is a framework which provides redundancy in distributed computing systems to speed up large-scale tasks. Although most existing works assume error-free scenarios, the link failures are common in current wired/wireless networks. In this paper, we consider the straggler problem in distributed computing systems with link failures, by modeling the links between the master node and worker nodes as packet erasure channels. We first analyze the latency in this setting using an (n, k) maximum distance separable (MDS) code. Then, we consider a setup where the number of retransmissions is limited due to the bandwidth constraint. By formulating practical optimization problems related to latency, bandwidth and probability of successful computation, we obtain achievable performance curves as a function of packet erasure probability.

In this paper, we propose bi-directional cooperative non-orthogonal multiple access (NOMA). Compared to conventional NOMA, the main contributions of bi-directional cooperative NOMA can be explained in two directions: 1) the proposed NOMA system is still efficient when the channel gains of scheduled users are almost the same and 2) the proposed NOMA system operates well without accurate channel-state information at the base station. In a two-user scenario, the closed-form ergodic capacity of bi-directional cooperative NOMA is derived, and it is proven to be better than those of other techniques. Based on the ergodic capacity, the algorithms to find optimal power allocations maximizing the user fairness and sum rate are presented. Outage probability is also derived, and we show that bi-directional cooperative NOMA achieves a power gain over uni-directional cooperative NOMA and a diversity gain over non-cooperative NOMA and orthogonal multiple access (OMA). We finally extend the bi-directional cooperative NOMA to a multiuser model. The analysis of ergodic capacity and outage probability in a two-user scenario is numerically verified. Also, simulation results show that bi-directional cooperative NOMA provides better data rates than the existing NOMA schemes as well as OMA in a multiuser scenario.

This paper investigates spectral shaping of multi-carrier-modulation waveforms based on combination of Nyquist windowing and subband filtering. The combined windowing/filtering allows simultaneous control on both subcarrier and subband spectra. When compared with the existing Nyquist windowing or subband filtering techniques under a fixed excess frame length constraint, the proposed scheme offers reduced sensitivity to carrier frequency and symbol timing offsets. Establishing an analytical tool based on the error spectrum stack consisting of the error signals evaluated at different signal delay positions, we explore the window-filter trade-off and provide the minimum interference power solution for given ranges of carrier frequency and symbol timing offsets. Our design targets low-latency applications having no provisions for high-precision synchronization and having potential need for spectrum aggregation.