In cell-free massive MIMO systems with multiple distributed access points (APs) serving multiple users over the same time-frequency resources, downlink beamforming is done through spatial precoding. Precoding vectors can be optimally designed to use the minimum downlink transmit power while satisfying a quality-of-service requirement for each user. However, existing centralized solutions to beamforming optimization pose challenges such as high communication overhead and processing delay. On the other hand, distributed approaches either require data exchange over the network that scales with the number of antennas or solve the problem for cellular systems where every user is served by only one AP. In this paper, we formulate a multi-user beamforming optimization problem to minimize the total transmit power subject to per-user SINR requirements and propose a distributed optimization algorithm based on the alternating direction method of multipliers (ADMM) to solve it. In our method, every AP solves an iterative optimization problem using its local channel state information. APs only need to share a real-valued vector of interference terms with the size of the number of users. Through simulation results, we demonstrate that our proposed algorithm solves the optimization problem within tens of ADMM iterations and can effectively satisfy per-user SINR constraints.
In realistic distributed optimization scenarios, individual nodes possess only partial information and communicate over bandwidth constrained channels. For this reason, the development of efficient distributed algorithms is essential. In our paper we addresses the challenge of unconstrained distributed optimization. In our scenario each node's local function exhibits strong convexity with Lipschitz continuous gradients. The exchange of information between nodes occurs through $3$-bit bandwidth-limited channels (i.e., nodes exchange messages represented by a only $3$-bits). Our proposed algorithm respects the network's bandwidth constraints by leveraging zoom-in and zoom-out operations to adjust quantizer parameters dynamically. We show that during our algorithm's operation nodes are able to converge to the exact optimal solution. Furthermore, we show that our algorithm achieves a linear convergence rate to the optimal solution. We conclude the paper with simulations that highlight our algorithm's unique characteristics.
Unmanned Aerial Vehicles (UAVs), due to their low cost and high flexibility, have been widely used in various scenarios to enhance network performance. However, the optimization of UAV trajectories in unknown areas or areas without sufficient prior information, still faces challenges related to poor planning performance and low distributed execution. These challenges arise when UAVs rely solely on their own observation information and the information from other UAVs within their communicable range, without access to global information. To address these challenges, this paper proposes the Qedgix framework, which combines graph neural networks (GNNs) and the QMIX algorithm to achieve distributed optimization of the Age of Information (AoI) for users in unknown scenarios. The framework utilizes GNNs to extract information from UAVs, users within the observable range, and other UAVs within the communicable range, thereby enabling effective UAV trajectory planning. Due to the discretization and temporal features of AoI indicators, the Qedgix framework employs QMIX to optimize distributed partially observable Markov decision processes (Dec-POMDP) based on centralized training and distributed execution (CTDE) with respect to mean AoI values of users. By modeling the UAV network optimization problem in terms of AoI and applying the Kolmogorov-Arnold representation theorem, the Qedgix framework achieves efficient neural network training through parameter sharing based on permutation invariance. Simulation results demonstrate that the proposed algorithm significantly improves convergence speed while reducing the mean AoI values of users. The code is available at https://github.com/UNIC-Lab/Qedgix.
In this paper, we extend the standard Attention in transformer by exploiting the consensus discrepancy from a distributed optimization perspective, referred to as AttentionX. It is noted that the primal-dual method of multipliers (PDMM) \cite{Zhang16PDMM} is designed to iteratively solve a broad class of distributed optimization problems over a pear-to-pear (P2P) network, where neighbouring nodes gradually reach consensus as specified by predefined linear edge-constraints in the optimization process. In particular, at each iteration of PDMM, each node in a network first performs information-gathering from neighbours and then performs local information-fusion. From a high-level point of view, the $KQ$-softmax-based weighted summation of $V$-representations in Attention corresponds information-gathering from neighbours while the feature-processing via the feed-forward network (FFN) in transformer corresponds to local information fusion. PDMM exploits the Lagrangian multipliers to capture the historical consensus discrepancy in the form of residual errors of the linear edge-constraints, which plays a crucial role for the algorithm to converge. Inspired by PDMM, we propose AttentionX to incorporate the consensus discrepancy in the output update-expression of the standard Attention. The consensus discrepancy in AttentionX refers to the difference between the weighted summation of $V$-representations and scaled $V$-representions themselves. Experiments on ViT and nanoGPT show promising performance.
This paper addresses distributed constrained multiobjective resource allocation problems (DCMRAPs) within multi-agent networks, where each agent has multiple, potentially conflicting local objectives, constrained by both local and global constraints. By reformulating the DCMRAP as a single-objective weighted $L_p$ problem, a distributed solution is enabled, which eliminates the need for predetermined weighting factors or centralized decision-making in traditional methods. Leveraging prescribed-time control and dynamic event-triggered mechanisms (ETMs), novel distributed algorithms are proposed to achieve Pareto optimality within a prescribed settling time through sampled communication. Using generalized time-based generators (TBGs), these algorithms provide more flexibility in optimizing accuracy and control smoothness without the constraints of initial conditions. Novel dynamic ETMs are designed to work with generalized TBGs to promote communication efficiency, which adjusts to both local error metrics and network-based disagreements. The Zeno behavior is excluded. Validated by Lyapunov analysis and simulations, our method demonstrates superior control performance and efficiency compared to existing methods, advancing distributed optimization in complex environments.
Federated Learning (FL) represents a paradigm shift in the field of machine learning, offering an approach for a decentralized training of models across a multitude of devices while maintaining the privacy of local data. However, the dynamic nature of FL systems, characterized by the ongoing incorporation of new clients with potentially diverse data distributions and computational capabilities, poses a significant challenge to the stability and efficiency of these distributed learning networks. The seamless integration of new clients is imperative to sustain and enhance the performance and robustness of FL systems. This paper looks into the complexities of integrating new clients into existing FL systems and explores how data heterogeneity and varying data distribution (not independent and identically distributed) among them can affect model training, system efficiency, scalability and stability. Despite these challenges, the integration of new clients into FL systems presents opportunities to enhance data diversity, improve learning performance, and leverage distributed computational power. In contrast to other fields of application such as the distributed optimization of word predictions on Gboard (where federated learning once originated), there are usually only a few clients in the production environment, which is why information from each new client becomes all the more valuable. This paper outlines strategies for effective client selection strategies and solutions for ensuring system scalability and stability. Using the example of images from optical quality inspection, it offers insights into practical approaches. In conclusion, this paper proposes that addressing the challenges presented by new client integration is crucial to the advancement and efficiency of distributed learning networks, thus paving the way for the adoption of Federated Learning in production environments.
Decentralized optimization has become a standard paradigm for solving large-scale decision-making problems and training large machine learning models without centralizing data. However, this paradigm introduces new privacy and security risks, with malicious agents potentially able to infer private data or impair the model accuracy. Over the past decade, significant advancements have been made in developing secure decentralized optimization and learning frameworks and algorithms. This survey provides a comprehensive tutorial on these advancements. We begin with the fundamentals of decentralized optimization and learning, highlighting centralized aggregation and distributed consensus as key modules exposed to security risks in federated and distributed optimization, respectively. Next, we focus on privacy-preserving algorithms, detailing three cryptographic tools and their integration into decentralized optimization and learning systems. Additionally, we examine resilient algorithms, exploring the design and analysis of resilient aggregation and consensus protocols that support these systems. We conclude the survey by discussing current trends and potential future directions.
In this paper, we study distributed prime-dual flows for multi-agent optimization with spatio-temporal compressions. The central aim of multi-agent optimization is for a network of agents to collaboratively solve a system-level optimization problem with local objective functions and node-to-node communication by distributed algorithms. The scalability of such algorithms crucially depends on the complexity of the communication messages, and a number of communication compressors for distributed optimization have recently been proposed in the literature. First of all, we introduce a general spatio-temporal compressor characterized by the stability of the resulting dynamical system along the vector field of the compressor. We show that several important distributed optimization compressors such as the greedy sparsifier, the uniform quantizer, and the scalarizer all fall into the category of this spatio-temporal compressor. Next, we propose two distributed prime-dual flows with the spatio-temporal compressors being applied to local node states and local error states, respectively, and prove (exponential) convergence of the node trajectories to the global optimizer for (strongly) convex cost functions. Finally, a few numerical examples are present to illustrate our theoretical results.
In this article, we focus on solving a class of distributed optimization problems involving $n$ agents with the local objective function at every agent $i$ given by the difference of two convex functions $f_i$ and $g_i$ (difference-of-convex (DC) form), where $f_i$ and $g_i$ are potentially nonsmooth. The agents communicate via a directed graph containing $n$ nodes. We create smooth approximations of the functions $f_i$ and $g_i$ and develop a distributed algorithm utilizing the gradients of the smooth surrogates and a finite-time approximate consensus protocol. We term this algorithm as DDC-Consensus. The developed DDC-Consensus algorithm allows for non-symmetric directed graph topologies and can be synthesized distributively. We establish that the DDC-Consensus algorithm converges to a stationary point of the nonconvex distributed optimization problem. The performance of the DDC-Consensus algorithm is evaluated via a simulation study to solve a nonconvex DC-regularized distributed least squares problem. The numerical results corroborate the efficacy of the proposed algorithm.
Distributed gradient descent algorithms have come to the fore in modern machine learning, especially in parallelizing the handling of large datasets that are distributed across several workers. However, scant attention has been paid to analyzing the behavior of distributed gradient descent algorithms in the presence of adversarial corruptions instead of random noise. In this paper, we formulate a novel problem in which adversarial corruptions are present in a distributed learning system. We show how to use ideas from (lazy) mirror descent to design a corruption-tolerant distributed optimization algorithm. Extensive convergence analysis for (strongly) convex loss functions is provided for different choices of the stepsize. We carefully optimize the stepsize schedule to accelerate the convergence of the algorithm, while at the same time amortizing the effect of the corruption over time. Experiments based on linear regression, support vector classification, and softmax classification on the MNIST dataset corroborate our theoretical findings.
This paper proposes a cooperative integrated sensing and communication network (Co-ISACNet) adopting hybrid beamforming (HBF) architecture, which improves both radar sensing and communication performance. The main contributions of this work are four-fold. First, we introduce a novel cooperative sensing method for the considered Co-ISACNet, followed by a comprehensive analysis of this method. This analysis mathematically verifies the benefits of Co-ISACNet and provides insightful design guidelines. Second, to show the benefits of Co-ISACNet, we propose to jointly design the HBF to maximize the network communication capacity while satisfying the constraint of beampattern similarity for radar sensing, which results in a highly dimensional and non-convex problem. Third, to facilitate the joint design, we propose a novel distributed optimization framework based on proximal gradient and alternating direction method of multipliers, namely PANDA. Fourth, we further adopt the proposed PANDA framework to solve the joint HBF design problem for the Co-ISACNet. By using the proposed PANDA framework, all access points (APs) optimize the HBF in parallel, where each AP only requires local channel state information and limited message exchange among the APs. Such framework reduces significantly the computational complexity and thus has pronounced benefits in practical scenarios. Simulation results verify the effectiveness of the proposed algorithm compared with the conventional centralized algorithm and show the remarkable performance improvement of radar sensing and communication by deploying Co-ISACNet.
We provide a communication- and computation-efficient method for distributed submodular optimization in robot mesh networks. Submodularity is a property of diminishing returns that arises in active information gathering such as mapping, surveillance, and target tracking. Our method, Resource-Aware distributed Greedy (RAG), introduces a new distributed optimization paradigm that enables scalable and near-optimal action coordination. To this end, RAG requires each robot to make decisions based only on information received from and about their neighbors. In contrast, the current paradigms allow the relay of information about all robots across the network. As a result, RAG's decision-time scales linearly with the network size, while state-of-the-art near-optimal submodular optimization algorithms scale cubically. We also characterize how the designed mesh-network topology affects RAG's approximation performance. Our analysis implies that sparser networks favor scalability without proportionally compromising approximation performance: while RAG's decision time scales linearly with network size, the gain in approximation performance scales sublinearly. We demonstrate RAG's performance in simulated scenarios of area detection with up to 45 robots, simulating realistic robot-to-robot (r2r) communication speeds such as the 0.25 Mbps speed of the Digi XBee 3 Zigbee 3.0. In the simulations, RAG enables real-time planning, up to three orders of magnitude faster than competitive near-optimal algorithms, while also achieving superior mean coverage performance. To enable the simulations, we extend the high-fidelity and photo-realistic simulator AirSim by integrating a scalable collaborative autonomy pipeline to tens of robots and simulating r2r communication delays. Our code is available at https://github.com/UM-iRaL/Resource-Aware-Coordination-AirSim.
Federated learning (FL) emerged as a paradigm designed to improve data privacy by enabling data to reside at its source, thus embedding privacy as a core consideration in FL architectures, whether centralized or decentralized. Contrasting with recent findings by Pasquini et al., which suggest that decentralized FL does not empirically offer any additional privacy or security benefits over centralized models, our study provides compelling evidence to the contrary. We demonstrate that decentralized FL, when deploying distributed optimization, provides enhanced privacy protection - both theoretically and empirically - compared to centralized approaches. The challenge of quantifying privacy loss through iterative processes has traditionally constrained the theoretical exploration of FL protocols. We overcome this by conducting a pioneering in-depth information-theoretical privacy analysis for both frameworks. Our analysis, considering both eavesdropping and passive adversary models, successfully establishes bounds on privacy leakage. We show information theoretically that the privacy loss in decentralized FL is upper bounded by the loss in centralized FL. Compared to the centralized case where local gradients of individual participants are directly revealed, a key distinction of optimization-based decentralized FL is that the relevant information includes differences of local gradients over successive iterations and the aggregated sum of different nodes' gradients over the network. This information complicates the adversary's attempt to infer private data. To bridge our theoretical insights with practical applications, we present detailed case studies involving logistic regression and deep neural networks. These examples demonstrate that while privacy leakage remains comparable in simpler models, complex models like deep neural networks exhibit lower privacy risks under decentralized FL.
In this work, we introduce the Resilient Projected Push-Pull (RP3) algorithm designed for distributed optimization in multi-agent cyber-physical systems with directed communication graphs and the presence of malicious agents. Our algorithm leverages stochastic inter-agent trust values and gradient tracking to achieve geometric convergence rates in expectation even in adversarial environments. We introduce growing constraint sets to limit the impact of the malicious agents without compromising the geometric convergence rate of the algorithm. We prove that RP3 converges to the nominal optimal solution almost surely and in the $r$-th mean for any $r\geq 1$, provided the step sizes are sufficiently small and the constraint sets are appropriately chosen. We validate our approach with numerical studies on average consensus and multi-robot target tracking problems, demonstrating that RP3 effectively mitigates the impact of malicious agents and achieves the desired geometric convergence.
Message passing graph neural networks (GNNs) would appear to be powerful tools to learn distributed algorithms via gradient descent, but generate catastrophically incorrect predictions when nodes update asynchronously during inference. This failure under asynchrony effectively excludes these architectures from many potential applications, such as learning local communication policies between resource-constrained agents in, e.g., robotic swarms or sensor networks. In this work we explore why this failure occurs in common GNN architectures, and identify "implicitly-defined" GNNs as a class of architectures which is provably robust to partially asynchronous "hogwild" inference, adapting convergence guarantees from work in asynchronous and distributed optimization, e.g., Bertsekas (1982); Niu et al. (2011). We then propose a novel implicitly-defined GNN architecture, which we call an energy GNN. We show that this architecture outperforms other GNNs from this class on a variety of synthetic tasks inspired by multi-agent systems, and achieves competitive performance on real-world datasets.
In distributed machine learning, efficient training across multiple agents with different data distributions poses significant challenges. Even with a centralized coordinator, current algorithms that achieve optimal communication complexity typically require either large minibatches or compromise on gradient complexity. In this work, we tackle both centralized and decentralized settings across strongly convex, convex, and nonconvex objectives. We first demonstrate that a basic primal-dual method, (Accelerated) Gradient Ascent Multiple Stochastic Gradient Descent (GA-MSGD), applied to the Lagrangian of distributed optimization inherently incorporates local updates, because the inner loops of running Stochastic Gradient Descent on the primal variable require no inter-agent communication. Notably, for strongly convex objectives, (Accelerated) GA-MSGD achieves linear convergence in communication rounds despite the Lagrangian being only linear in the dual variables. This is due to a structural property where the dual variable is confined to the span of the coupling matrix, rendering the dual problem strongly concave. When integrated with the Catalyst framework, our approach achieves nearly optimal communication complexity across various settings without the need for minibatches.
We study the distributed optimization problem over a graphon with a continuum of nodes, which is regarded as the limit of the distributed networked optimization as the number of nodes goes to infinity. Each node has a private local cost function. The global cost function, which all nodes cooperatively minimize, is the integral of the local cost functions on the node set. We propose stochastic gradient descent and gradient tracking algorithms over the graphon. We establish a general lemma for the upper bound estimation related to a class of time-varying differential inequalities with negative linear terms, based upon which, we prove that for both kinds of algorithms, the second moments of the nodes' states are uniformly bounded. Especially, for the stochastic gradient tracking algorithm, we transform the convergence analysis into the asymptotic property of coupled nonlinear differential inequalities with time-varying coefficients and develop a decoupling method. For both kinds of algorithms, we show that by choosing the time-varying algorithm gains properly, all nodes' states achieve $\mathcal{L}^{\infty}$-consensus for a connected graphon. Furthermore, if the local cost functions are strongly convex, then all nodes' states converge to the minimizer of the global cost function and the auxiliary states in the stochastic gradient tracking algorithm converge to the gradient value of the global cost function at the minimizer uniformly in mean square.
In this paper, we study the problem of utility maximization in the uplink of vehicle-to-infrastructure communication systems. The studied scenarios consider four practical aspects of mobile radio communication links: i) Interference between adjacent channels, ii) interference between roadside units along the way, iii) fast and slow channel fadings, and iv) Doppler shift effects. We present first the system model for the IEEE 802.11p standard, which considers a communication network between vehicles and roadside infrastructure. Next, we formulate the problem of utility maximization in the network, and propose a distributed optimization scheme. This distributed scheme is based on a two-loop feedback configuration, where an outer-loop establishes the optimal signal to interference-noise ratio (SINR) that maximizes the utility function per vehicle and defines a quality-of-service objective. Meanwhile, inner-control loops adjust the transmission power to achieve this optimal SINR reference in each vehicle node regardless of interference, time-varying channel profiles and network latency. The computation complexity of the distributed utility maximization scheme is analyzed for each feedback loop. Simulation results indicate that the proposed scheme reaches the objective SINRs that maximize utility and improve energy efficiency in the network with a low time cost. The results also show that the maximum utility is consistently achieved for different propagation scenarios inside the vehicular communication network.
Training LLMs relies on distributed implementations using multiple GPUs to compute gradients in parallel with sharded optimizers. However, synchronizing gradients in data parallel setups introduces communication overhead that grows with the number of workers, limiting parallelization efficiency. Local optimization algorithms reduce communications but incur high memory costs as they prevent optimizer state sharding, hindering scalability. To address this, we propose \textbf{AC}cumulate while \textbf{CO}mmunicate (\acco), a memory-efficient optimization algorithm for distributed LLM training. By synchronizing delayed gradients while computing new ones, \acco~reduces GPU idle time and supports heterogeneous hardware. To mitigate the convergence issues caused by delayed updates, we introduce a novel technique ensuring training dynamics align with standard distributed optimization. Compared to ZeRO-1, our approach is significantly faster and scales effectively across heterogeneous hardware.
Decentralized strategies are of interest for learning from large-scale data over networks. This paper studies learning over a network of geographically distributed nodes/agents subject to quantization. Each node possesses a private local cost function, collectively contributing to a global cost function, which the considered methodology aims to minimize. In contrast to many existing papers, the information exchange among nodes is log-quantized to address limited network-bandwidth in practical situations. We consider a first-order computationally efficient distributed optimization algorithm (with no extra inner consensus loop) that leverages node-level gradient correction based on local data and network-level gradient aggregation only over nearby nodes. This method only requires balanced networks with no need for stochastic weight design. It can handle log-scale quantized data exchange over possibly time-varying and switching network setups. We study convergence over both structured networks (for example, training over data-centers) and ad-hoc multi-agent networks (for example, training over dynamic robotic networks). Through experimental validation, we show that (i) structured networks generally result in a smaller optimality gap, and (ii) log-scale quantization leads to a smaller optimality gap compared to uniform quantization.