Inertial Proximal Alternating Linearized Minimization (iPALM) for Nonconvex and Nonsmooth Problems

Abstract

In this paper we study nonconvex and nonsmooth optimization problems with semi-algebraic data, where the variables vector is split into several blocks of variables. The problem consists of one smooth function of the entire variables vector and the sum of nonsmooth functions for each block separately. We analyze an inertial version of the Proximal Alternating Linearized Minimization (PALM) algorithm and prove its global convergence to a critical point of the objective function at hand. We illustrate our theoretical findings by presenting numerical experiments on blind image deconvolution, on sparse non-negative matrix factorization and on dictionary learning, which demonstrate the viability and effectiveness of the proposed method.

References in corpus (1)

• A globally convergent algorithm for nonconvex optimization based on block coordinate update

Cited by in corpus (54)

• Adaptive Diffusion Priors for Accelerated MRI Reconstruction
• Blind Image Fusion for Hyperspectral Imaging with the Directional Total Variation
• The Practicality of Stochastic Optimization in Imaging Inverse Problems
• Directional Sinogram Inpainting for Limited Angle Tomography
• Learning a Variational Network for Reconstruction of Accelerated MRI Data
• Acceleration and global convergence of a first-order primal--dual method for nonconvex problems
• Stable Deep MRI Reconstruction using Generative Priors
• A Framework of Inertial Alternating Direction Method of Multipliers for Non-Convex Non-Smooth Optimization
• 3D Fluid Flow Estimation with Integrated Particle Reconstruction
• Optimization on Spheres: Models and Proximal Algorithms with Computational Performance Comparisons
• GEASI: Geodesic-based Earliest Activation Sites Identification in cardiac models
• A convex variational model for learning convolutional image atoms from incomplete data
• The basins of attraction of the global minimizers of non-convex inverse problems with low-dimensional models in infinite dimension
• Deep, Deep Learning with BART
• Beyond Alternating Updates for Matrix Factorization with Inertial Bregman Proximal Gradient Algorithms
• Variational 3D-PIV with Sparse Descriptors
• Inertial Stochastic PALM (iSPALM) and Applications in Machine Learning
• Block Alternating Bregman Majorization Minimization with Extrapolation
• Regularization of Inverse Problems via Time Discrete Geodesics in Image Spaces
• SPRING: A fast stochastic proximal alternating method for non-smooth non-convex optimization
• Multi-block Bregman proximal alternating linearized minimization and its application to orthogonal nonnegative matrix factorization
• Matrix cofactorization for joint spatial-spectral unmixing of hyperspectral images
• An Inertial Block Majorization Minimization Framework for Nonsmooth Nonconvex Optimization
• Non-ergodic Convergence Analysis of Heavy-Ball Algorithms
• A Refined Inertial DC Algorithm for DC Programming
• Bregman Proximal Framework for Deep Linear Neural Networks
• A reciprocity-aware, low-rank regularization for multidimensional deconvolution
• Short-and-Sparse Deconvolution -- A Geometric Approach
• An Optimal Control Approach to Early Stopping Variational Methods for Image Restoration
• Consistency and Computation of Regularized MLEs for Multivariate Hawkes Processes
• Explicit Diffusion of Gaussian Mixture Model Based Image Priors
• LAMA: Stable Dual-Domain Deep Reconstruction For Sparse-View CT
• Model Function Based Conditional Gradient Method with Armijo-like Line Search
• Compressed Sensing Microscopy with Scanning Line Probes
• LAMA-Net: A Convergent Network Architecture for Dual-Domain Reconstruction
• Inertial Block Proximal Methods for Non-Convex Non-Smooth Optimization
• Group-sparse SVD Models and Their Applications in Biological Data
• Unifying abstract inexact convergence theorems and block coordinate variable metric iPiano
• Bregman Proximal Gradient Algorithm with Extrapolation for a class of Nonconvex Nonsmooth Minimization Problems
• Block Majorization Minimization with Extrapolation and Application to $\beta$-NMF
• PCA Reduced Gaussian Mixture Models with Applications in Superresolution
• PALMNUT: An Enhanced Proximal Alternating Linearized Minimization Algorithm with Application to Separate Regularization of Magnitude and Phase
• Proximal Residual Flows for Bayesian Inverse Problems
• An accelerated proximal iterative hard thresholding method for ${\ell }_0$ minimization
• Structured Sparse Non-negative Matrix Factorization with L20-Norm for scRNA-seq Data Analysis
• Inertial Bregman Proximal Gradient Algorithm For Nonconvex Problem with Smooth Adaptable Property
• Deep learning based dictionary learning and tomographic image reconstruction
• Convergence of Random Reshuffling Under The Kurdyka-Łojasiewicz Inequality
• Time series forecasting from partial observations via Non-negative Matrix Factorization
• Consistent Approximation of Interpolating Splines in Image Metamorphosis
• Non-ergodic Complexity of Convex Proximal Inertial Gradient Descents
• Proximal Algorithms for Smoothed Online Convex Optimization with Predictions
• Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for ${\ell }_1$-Norm Principal Component Analysis
• Inertial Proximal Deep Learning Alternating Minimization for Efficient Neutral Network Training