Distributed Primal-Dual Perturbation Algorithm Over Unbalanced Directed Networks

Sakuma, Hiroaki; Hayashi, Naoki; Takai, Shigemasa

doi:10.1109/access.2021.3082537

Cited by 10 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, Proposition 1 states that the convergence rate of the estimated primal solution to the optimal solution is sublinear. This shows that, despite the less local communication between service providers, the proposed event‐driven distributed algorithm can achieve comparable convergence performance to the existing time‐driven algorithms [20–27].…”

Section: Resultsmentioning

confidence: 98%

“…An approach to resolve the scalability issue is to use a distributed optimization algorithm for multi-agent systems, which has been gaining significant attention [20][21][22][23][24][25][26][27]. However, conventional distributed optimization methods require communication between service providers at each algorithm iteration.…”

Section: Related Workmentioning

confidence: 99%

“…The optimization problem (31) can be solved in a distributed and coordinated manner by distributed optimization algorithms [21,24,25]. Nevertheless, in most of the existing distributed optimization algorithms, at each iteration of the algorithm, each service provider should communicate with other providers, which may result in unnecessary information exchanges and congestion of the communication network.…”

Section: Reformulation Into Distributed Rebalancing Optimizationmentioning

confidence: 99%

See 2 more Smart Citations

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Hayashi

Sakurama

2022

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

This study presents a cooperative car‐sharing system in which various service providers jointly operate a sharing service. A distributed algorithm is proposed for rebalancing control to reduce deadhead vehicles as much as possible under the constraints of dispatchable vehicles and parking slots at car stations. Each provider cooperatively searches for the optimal solution by exchanging the estimated values of the optimal dual solutions over the communication networks between providers. Event‐driven communication is considered in which information exchange with the neighboring providers is performed only when the amount of change of the estimated optimal dual solution exceeds a threshold. Numerical instances explain that the estimations of all service providers meet the optimal rebalancing solutions.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Related Workmentioning

confidence: 99%

Section: Reformulation Into Distributed Rebalancing Optimizationmentioning

confidence: 99%

See 1 more Smart Citation

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Hayashi

Sakurama

2022

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly to Theorem 1, we give the decoupling decomposition of ( 29) in following theorem. Theorem 3: For the optimization problem (29), if there is a local cost function J i (y, α u , β v , σ) for every i ∈ N and for then G = {N , {Y i }, {J i }} is a potential game, and ( 29) can be decoupled into n sub-optimization problems…”

Section: A Decoupling Design Under Penalty Methodsmentioning

confidence: 99%

Cooperative Optimization With Globally Coupled Cost Function and Coupled Constraints

Li,

Wang,

Meng

2024

IEEE Access

View full text Add to dashboard Cite

We study the cooperative optimization problem (COP) of multi-agent systems with globally coupled cost function and coupled constraints, and design a distributed computing framework combining potential game theory underlying geometric projection. This design framework has the advantage of being able to solve the COP with globally coupled cost function and coupled constraints in distributed way. Firstly, the studied COP with coupled constraints is converted to an unconstrained one by using barrier and penalty methods, respectively, and then the coupled cost function of the unconstrained COP with n variables to be optimized is decoupled by projecting it to n hyperplanes, and n decoupled sub-optimization problems are established. Underlying this design, we exploit an equivalently changing relationship during the optimizing process between each decoupled cost function and the original global function in a fixed communication topology, which forms a potential game, and derive that the optimal solution of the COP is equivalent with Nash equilibrium of the potential game. The obtained sub-optimization problems can be solved in distributed manners and two improved distributed gradient algorithms are proposed. Finally, the distributed design is applied to the economic dispatch problem in power system to verify the superiority of the proposed algorithms.INDEX TERMS Cooperative optimization, coupled inequality constraint, potential game, barrier method, penalty method.

show abstract

“…On the other hand, the proposed method is based on the row stochastic property of the weight matrix. Because algorithms with row stochasticity can be implemented with the information received from inneighbor agents [41,42], the proposed method has an advantage in distributed settings compared with push-sum-based methods. Although the dual averaging algorithm in [40] does not require column stochasticity, a coupling inequality constraint is not considered.…”

Section: Introductionmentioning

confidence: 99%

Constrained distributed online convex optimization with bandit feedback for unbalanced digraphs

Tada,

Hayashi,

Takai

2023

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

In this study, a distributed primal‐dual bandit feedback method for online convex optimization with time‐varying coupled inequality constraints on unbalanced directed graphs is proposed. A multiagent network is considered in which agents exchange the estimations of the dual optimizer and the scaling variable with their neighbors. The scaling variable is used to resolve the bias of the estimations caused by a directed communication network. Each agent does not have prior knowledge of the loss function, and its value at a queried point is sequentially disclosed to each agent. Each agent performs a projected subgradient‐based primal‐dual algorithm to estimate the optimal solution. It is confirmed that both the expected dynamic regret of the loss function and the cumulative error of the constraint violation achieve sublinearity using the proposed online algorithm with the two‐point bandit feedback.

show abstract

Distributed Primal-Dual Perturbation Algorithm Over Unbalanced Directed Networks

Abstract: Date of publication xxxx , , date of current version xxxx , ****.

Cited by 10 publications

References 33 publications

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Cooperative Optimization With Globally Coupled Cost Function and Coupled Constraints

Constrained distributed online convex optimization with bandit feedback for unbalanced digraphs

Contact Info

Product

Resources

About

Distributed Primal-Dual Perturbation Algorithm Over Unbalanced Directed Networks

Abstract: Date of publication xxxx **, ****, date of current version xxxx **, ****.

Cited by 10 publications

References 33 publications

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Communication‐aware distributed rebalancing for cooperative car‐sharing service

Cooperative Optimization With Globally Coupled Cost Function and Coupled Constraints

Constrained distributed online convex optimization with bandit feedback for unbalanced digraphs

Contact Info

Product

Resources

About

Abstract: Date of publication xxxx , , date of current version xxxx , ****.