Distributed Subgradient Methods for Multi-Agent Optimization
Angelia Nedić, Asuman Ozdaglar · IEEE Transactions on Automatic Control · 2009
We study a distributed computation model for optimizing a sum of convex objective functions corresponding to multiple agents. For solving this (not necessarily smooth) optimization problem, we consider a subgradient method that is distributed among the agents. The method involves every agent minimizing his/her own objective function while exchanging information locally with other agents in the network over a time-varying topology. We provide convergence results and convergence rate estimates for the subgradient method. Our convergence rate results explicitly characterize the tradeoff between a desired accuracy of the generated approximate optimal solutions and the number of iterations needed to achieve the accuracy.
Cited by 3,690 works. doi:10.1109/tac.2008.2009515
Authors
References
- Consensus Problems in Networks of Agents With Switching Topology and Time-Delays
- Coordination of groups of mobile autonomous agents using nearest neighbor rules
- Parallel and Distributed Computation: Numerical Methods
- Novel Type of Phase Transition in a System of Self-Driven Particles
- Rate control for communication networks: shadow prices, proportional fairness and stability
- Rate control for communication networks: shadow prices, proportional fairness and stability
- Distributed asynchronous deterministic and stochastic gradient optimization algorithms
- Minimization Methods for Non-Differentiable Functions
- Parallel and distributed computation
- Optimization flow control. I. Basic algorithm and convergence