litlas
ログインアカウント作成

A General Analysis of the Convergence of ADMM

著者: Robert Nishihara, Laurent Lessard, Benjamin Recht, Andrew K. Packard, Michael I. Jordan - International Conference on Machine Learning, ICML 2015 被引用: 339

We provide a new proof of the linear convergence of the alternating direction method of multipliers (ADMM) when one of the objective terms is strongly convex. Our proof is based on a framework for analyzing optimization algorithms introduced in Lessard et al. (2014), reducing algorithm convergence to verifying the stability of a dynamical system. This approach generalizes a number of existing results and obviates any assumptions about specific choices of algorithm parameters. On a numerical example, we demonstrate that minimizing the derived bound on the convergence rate provides a practical approach to selecting algorithm parameters for particular ADMM instances. We complement our upper bound by constructing a nearly-matching lower bound on the worst-case rate of convergence.

On the Douglas-Rachfordsplitting method and th…On the Douglas-Rachford splitting method and the proximal point algorithm for maximal monotone operatorsParallel alternatingdirection multiplier…Parallel alternating direction multiplier decomposition of convex programsDistributed Optimizationand Statistical Learnin…Distributed Optimization and Statistical Learning via the Alternating Direction Method of MultipliersAn Alternating DirectionMethod for Dual MAP LP…An Alternating Direction Method for Dual MAP LP RelaxationOn the Global and LinearConvergence of the…On the Global and Linear Convergence of the Generalized Alternating Direction Method of MultipliersEfficient DistributedLinear Classification…Efficient Distributed Linear Classification Algorithms via the Alternating Direction Method of MultipliersOnline AlternatingDirection MethodOnline Alternating Direction MethodDiagonal scaling inDouglas-Rachford…Diagonal scaling in Douglas-Rachford splitting and ADMMFast AlternatingDirection Optimization…Fast Alternating Direction Optimization MethodsOn the linearconvergence of the…On the linear convergence of the alternating direction method of multipliersConvergence RateAnalysis of Several…Convergence Rate Analysis of Several Splitting SchemesFaster Convergence Ratesof Relaxed…Faster Convergence Rates of Relaxed Peaceman-Rachford and ADMM Under Regularity AssumptionsTight linear convergencerate bounds for…Tight linear convergence rate bounds for Douglas-Rachford splitting and ADMMFast-and-LightStochastic ADMMFast-and-Light Stochastic ADMMConvergence Rate ofDistributed ADMM Over…Convergence Rate of Distributed ADMM Over NetworksAdaptive ADMM withSpectral Penalty…Adaptive ADMM with Spectral Penalty Parameter SelectionAdaptive Consensus ADMMfor Distributed…Adaptive Consensus ADMM for Distributed OptimizationAdaptive Relaxed ADMM:Convergence Theory and…Adaptive Relaxed ADMM: Convergence Theory and Practical ImplementationDykstra's Algorithm,ADMM, and Coordinate…Dykstra's Algorithm, ADMM, and Coordinate Descent: Connections, Insights, and ExtensionsAdaptive ADMM forDistributed AC Optimal…Adaptive ADMM for Distributed AC Optimal Power FlowParameter Selection andPreconditioning for a…Parameter Selection and Preconditioning for a Graph Form SolverOn Systems andAlgorithms for…On Systems and Algorithms for Distributed Machine LearningAccelerated VarianceReduction Stochastic…Accelerated Variance Reduction Stochastic ADMM for Large-Scale Machine LearningQuantifying theasymptotic linear…Quantifying the asymptotic linear convergence speed of Anderson Acceleration applied to ADMMA General Analysis ofthe Convergence of ADMMA General Analysis of the Convergence of ADMM過去の参考文献中心の論文この論文を引用する論文古い新しい

ノードをクリックするとフォーカスを固定、空白をクリックすると本論文に戻ります。ホバーで一時的にプレビューできます。各ノードのページはタイトルから開けます。