A Proof of Convergence For the Alternating Direction Method of Multipliers Applied to Polyhedral-Constrained Functions
arXiv:1112.2295
Abstract
We give a general proof of convergence for the Alternating Direction Method of Multipliers (ADMM). ADMM is an optimization algorithm that has recently become very popular due to its capabilities to solve large-scale and/or distributed problems. We prove that the sequence generated by ADMM converges to an optimal primal-dual optimal solution. We assume the functions f and g, defining the cost f(x) + g(y), are real-valued, but constrained to lie on polyhedral sets X and Y. Our proof is an extension of the proofs from [Bertsekas97, Boyd11].
Cited by in corpus (10)
- D-ADMM: A Communication-Efficient Distributed Algorithm For Separable Optimization
- Distributed Optimization With Local Domains: Applications in MPC and Network Flows
- Multi-Agent Reinforcement Learning via Distributed MPC as a Function Approximator
- Distributed Training of Structured SVM
- Communication-Efficient Algorithms For Distributed Optimization
- Anomaly Detection via Graphical Lasso
- An ADMM-based MIQP platform for the EV aggregation management
- Disease Prediction based on Functional Connectomes using a Scalable and Spatially-Informed Support Vector Machine
- Distributed Event Localization via Alternating Direction Method of Multipliers
- Local-Aggregate Modeling for Big-Data via Distributed Optimization: Applications to Neuroimaging