The effect of decentralized resource allocation in network-centric warfare

Rhee, Seung Hyong; Kim, Hwa‐Sung; Sohn, Seung-Won

doi:10.1109/icoin.2012.6164447

Cited by 8 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the existing literature of decentralized learning are focused on the standard loss minimization formulation [41,43], i.e., min x∈R 𝑑 𝑓 (x), where 𝑓 (•) is the objective loss function and x denotes the global model parameters to be learned, and 𝑑 is the model dimension. In the literature, a wide range of machine learning applications can be modeled by the standard decentralized loss minimization formulation (e.g., robotic network [15,32], network resource allocation [12,36], power networks [2,7]). Some recent works, [20,21,25,44] studied decentralized min-max optimization problems, i.e., min x∈R 𝑑 1 max y∈R 𝑑 2 𝑓 (x, y), which are a special case (with same outer and inner level objective) of bilevel optimization problems.…”

Section: Related Workmentioning

confidence: 99%

INTERACT: Achieving Low Sample and Communication Complexities in Decentralized Bilevel Learning over Networks

Liu¹,

Zhang²,

Khanduri³

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years, decentralized bilevel optimization problems have received increasing attention in the networking and machine learning communities thanks to their versatility in modeling decentralized learning problems over peer-to-peer networks (e.g., multi-agent meta-learning, multi-agent reinforcement learning, personalized training, and Byzantine-resilient learning). However, for decentralized bilevel optimization over peer-to-peer networks with limited computation and communication capabilities, how to achieve low sample and communication complexities are two fundamental challenges that remain under-explored so far. In this paper, we make the first attempt to investigate the class of decentralized bilevel optimization problems with nonconvex and strongly-convex structure corresponding to the outer and inner subproblems, respectively. Our main contributions in this paper are two-fold: i) We first propose a deterministic algorithm called INTERACT (innergradient-descent-outer-tracked-gradient) that requires the sample complexity of O (𝑛𝜖 −1 ) and communication complexity of O (𝜖 −1 ) to solve the bilevel optimization problem, where 𝑛 and 𝜖 > 0 are the number of samples at each agent and the desired stationarity gap, respectively. ii) To relax the need for full gradient evaluations in each iteration, we propose a stochastic variance-reduced version of INTERACT (SVR-INTERACT), which improves the sample complexity to O ( √ 𝑛𝜖 −1 ) while achieving the same communication complexity as the deterministic algorithm. To our knowledge, this work is the first that achieves both low sample and communication complexities for solving decentralized bilevel optimization problems over networks. Our numerical experiments also corroborate our theoretical findings.

show abstract

Section: Related Workmentioning

confidence: 99%

INTERACT: Achieving Low Sample and Communication Complexities in Decentralized Bilevel Learning over Networks

Liu¹,

Zhang²,

Khanduri³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…These real-world limitations have spawned the rapid development of decentralized learning over edge networks in recent years, which can leverage highly flexible peer-to-peer edge computing networks with arbitrary topologies Nedic and Ozdaglar [2009], Lian et al [2017]. Also, thanks to the resilience to single-pointof-failure, data privacy, and simple implementations, decentralized learning has attracted growing interest recently, and has found various science and engineering applications, such as distributedrobotics control Ren et al [2007], Zhou and Roumeliotis [2011] and network resource allocation Jiang et al [2018], Rhee et al [2012], such as dictionary learning Chen et al [2014], multi-agent systems Cao et al [2012], Zhou and Roumeliotis [2011], multi-task learning Wang et al [2018], Zhang et al [2019], and information retrieval Ali and Van Stam [2004].…”

Section: Introductionmentioning

confidence: 99%

PRECISION: Decentralized Constrained Min-Max Learning with Low Communication and Sample Complexities

Liu¹,

Zhang²,

Liu³

2023

Preprint

View full text Add to dashboard Cite

Recently, min-max optimization problems have received increasing attention due to their wide range of applications in machine learning (ML). However, most existing min-max solution techniques are either single-machine or distributed algorithms coordinated by a central server. In this paper, we focus on the decentralized min-max optimization for learning with domain constraints, where multiple agents collectively solve a nonconvex-strongly-concave min-max saddle point problem without coordination from any server. Decentralized min-max optimization problems with domain constraints underpins many important ML applications, including multi-agent ML fairness assurance, and policy evaluations in multi-agent reinforcement learning. We propose an algorithm called PRECISION (proximal gradient-tracking and stochastic recursive variance reduction) that enjoys a convergence rate of O(1/T ), where T is the maximum number of iterations. To further reduce sample complexity, we propose PRECISION + with an adaptive batch size technique. We show that the fast O(1/T ) convergence of PRECISION and PRECISION + to an -stationary point imply O( −2 ) communication complexity and O(m √ n −2 ) sample complexity, where m is the number of agents and n is the size of dataset at each agent. To our knowledge, this is the first work that achieves O( −2 ) in both sample and communication complexities in decentralized min-max learning with domain constraints. Our experiments also corroborate the theoretical results.

show abstract

“…The completion of missions often requires detailed decomposition and interpretation, and is broken down into a series of specific tasks or subtasks to perform [23]. In [24], a distributed decision-making capability resulting in near-optimal weapon-target assignments for formations of unmanned combat vehicles was proposed. The decision-making is based on a modified version of the cross-entropy method distributed over the formations.…”

Section: Introductionmentioning

confidence: 99%

Mission-Oriented Scheme Generation Method for Weapon System of Systems

Zhao

Ding

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The scheme for weapon system of systems (WSoS) is a combination of equipment systems that supports the mission completion. Modern warfare combat missions tend to be complex, short-lived, and costly, which poses new challenges to the generation of the schema of WSoS. The structures of both missions and WSoS are analyzed in this paper. The subtask decomposition relationship of the mission and the system decomposition relationship of the equipment system are defined. A fuzzy quantitative description of the system function implementation is formed in terms of supportiveness, time and cost. On this basis, the mission-oriented WSoS scheme is transformed into a fuzzy weighted network, and the scheme generation algorithm and scheme evaluation index are proposed. Finally, the feasibility of the proposed approach is demonstrated using an illustrative example.

show abstract

The effect of decentralized resource allocation in network-centric warfare

Cited by 8 publications

References 4 publications

INTERACT: Achieving Low Sample and Communication Complexities in Decentralized Bilevel Learning over Networks

INTERACT: Achieving Low Sample and Communication Complexities in Decentralized Bilevel Learning over Networks

PRECISION: Decentralized Constrained Min-Max Learning with Low Communication and Sample Complexities

Mission-Oriented Scheme Generation Method for Weapon System of Systems

Contact Info

Product

Resources

About