A Unified Algorithmic Framework for Block-Structured Optimization Involving Big Data
arXiv:1511.02746
Abstract
This article presents a powerful algorithmic framework for big data optimization, called the Block Successive Upper bound Minimization (BSUM). The BSUM includes as special cases many well-known methods for analyzing massive data sets, such as the Block Coordinate Descent (BCD), the Convex-Concave Procedure (CCCP), the Block Coordinate Proximal Gradient (BCPG) method, the Nonnegative Matrix Factorization (NMF), the Expectation Maximization (EM) method and so on. In this article, various features and properties of the BSUM are discussed from the viewpoint of design flexibility, computational efficiency, parallel/distributed implementation and the required communication overhead. Illustrative examples from networking, signal processing and machine learning are presented to demonstrate the practical performance of the BSUM framework
References in corpus (18)
- Decomposition by Partial Linearization: Parallel Optimization of Multi-Agent Systems
- Guaranteed Matrix Completion via Non-convex Factorization
- Convex Optimization for Big Data
- Linear Precoding in Cooperative MIMO Cellular Networks with Limited Coordination Clusters
- Distributed Coordinate Descent Method for Learning with Big Data
- Parallel Selective Algorithms for Big Data Optimization
- An Asynchronous Parallel Stochastic Coordinate Descent Algorithm
- Near-optimal RNA-Seq quantification
- Stochastic Majorization-Minimization Algorithms for Large-Scale Optimization
- Rank regularization and Bayesian inference for tensor completion and extrapolation
- Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection
- A Block Successive Upper Bound Minimization Method of Multipliers for Linearly Constrained Convex Optimization
- Feature Clustering for Accelerating Parallel Coordinate Descent
- Accelerated, Parallel and Proximal Coordinate Descent
- An Accelerated Proximal Coordinate Gradient Method and its Application to Regularized Empirical Risk Minimization
- Iteration Complexity Analysis of Block Coordinate Descent Methods
- Large-scale randomized-coordinate descent methods with non-separable linear constraints
- Decomposition by Successive Convex Approximation: A Unifying Approach for Linear Transceiver Design in Heterogeneous Networks
Cited by in corpus (10)
- Hybrid Block Successive Approximation for One-Sided Non-Convex Min-Max Problems: Algorithms and Applications
- Non-convex Min-Max Optimization: Applications, Challenges, and Recent Theoretical Advances
- Joint Source-Relay Design for Full--Duplex MIMO AF Relay Systems
- Block-Term Tensor Decomposition: Model Selection and Computation
- Zeroth Order Nonconvex Multi-Agent Optimization over Networks
- Energy-Efficient UAV Backscatter Communication with Joint Trajectory Design and Resource Optimization
- Block Coordinate Descent Only Converge to Minimizers
- On the Pervasiveness of Difference-Convexity in Optimization and Statistics
- Gradient-Free Multi-Agent Nonconvex Nonsmooth Optimization
- Convergence Analysis and Design of Multi-block ADMM via Switched Control Theory