Distributed classification learning based on nonlinear vector support machines for switching networks

Wang, Yinghui; Lin, Peng; Qin, Huashu

doi:10.14736/kyb-2017-4-0595

Cited by 5 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the Nash equilibrium is (1/(N + 1), • • • , 1/(N + 1)). In this aggregative game, the local payoff of each user i is f i = (1/(N + 1)) 2 and the global payoff is…”

Section: Problem Formulationsmentioning

confidence: 99%

“…However, If all agents cooperate to maximize the global payoff function, the optimal solution is computed as x * = (1/2N, • • • , 1/2N ). In this setting, the local payoff of each user i is f i = 1/4N and the global payoff is 2 . This example indicates that all agents will perform better in a cooperative manner compared with the aggregative game in a noncooperative manner, which motivates us to study the aggregative optimization as (1).…”

Section: Problem Formulationsmentioning

confidence: 99%

“…Distributed optimization has gained much research attention due to its wide applications in multi-agent network systems, such as resource allocation [1], machine learning [2,3] and cloud computing [4]. In distributed optimization, each agent only uses local data and transmits information with its neighbors to minimize a global cost function cooperatively.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Distributed aggregative optimization with quantization communication

Chen¹,

Shen²

2022

Preprint

View full text Add to dashboard Cite

In this paper, we focus on an aggregative optimization problem under communication bottleneck. The aggregative optimization is to minimize the sum of local cost functions. Each cost function depends on not only local state variables but also the sum of functions of global state variables. The goal is to solve the aggregative optimization problem through distributed computation and local efficient communication over a network of agents without a central coordinator. Using the variable tracking method to seek the global state variables and the quantization scheme to reduce the communication cost spent in the optimization process, we develop a novel distributed quantized algorithm, called D-QAGT, to track the optimal variables with finite bits communication. Although quantization may lose transmitting information, our algorithm can still achive the exact optimal solution with linear convergence rate. Simulation experiments on an optimal placement problem is carried out to verify the correctness of the theoretical results.

show abstract

“…Thus, the Nash equilibrium is (1/(N + 1), • • • , 1/(N + 1)). In this aggregative game, the local payoff of each user i is f i = (1/(N + 1)) 2 and the global payoff is…”

Section: Problem Formulationsmentioning

confidence: 99%

Section: Problem Formulationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distributed aggregative optimization with quantization communication

Chen¹,

Shen²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Two-subnetwork zero-sum games are an important class of distributed non-cooperative games that have found a wide range of applications, including signal/image processing [4], statistical learning [29,35,40], formation control [21,25,44], and resource allocation [11,20,23,41]. In a two-subnetwork zero-sum game (see [24,34]), agents in one subnetwork cooperate to minimize their payoff function using local information exchange, and agents in the other subnetwork try to maximize the same payoff function.…”

Section: Introductionmentioning

confidence: 99%

Distributed accelerated Nash equilibrium learning for two-subnetwork zero-sum game with bilinear coupling

Zeng¹,

Dou²,

Cui³

2023

Kybernetika

View full text Add to dashboard Cite

This paper proposes a distributed accelerated first-order continuous-time algorithm for O(1/t 2 ) convergence to Nash equilibria in a class of two-subnetwork zero-sum games with bilinear couplings. First-order methods, which only use subgradients of functions, are frequently used in distributed/parallel algorithms for solving large-scale and big-data problems due to their simple structures. However, in the worst cases, first-order methods for two-subnetwork zero-sum games often have an asymptotic or O(1/t) convergence. In contrast to existing timeinvariant first-order methods, this paper designs a distributed accelerated algorithm by combining saddle-point dynamics and time-varying derivative feedback techniques. If the parameters of the proposed algorithm are suitable, the algorithm owns O(1/t 2 ) convergence in terms of the duality gap function without any uniform or strong convexity requirement. Numerical simulations show the efficacy of the algorithm.

show abstract

“…As a result, distributed algorithms have attracted much research attention. Particularly, distributed optimization, which agents over the network cooperately seeks a global optimal solution, has become more and more popular [3,13,14,16,17].…”

Section: Introductionmentioning

confidence: 99%