On the Convergence of the Stochastic Primal-Dual Hybrid Gradient for Convex Optimization

Stochastic Primal-Dual Hybrid Gradient (SPDHG) was proposed by Chambolle et al. (2018) and is a practical tool to solve nonsmooth large-scale optimization problems. In this paper we prove its almost sure convergence for convex but not necessarily strongly convex functionals. The proof makes use of a classical supermartingale result, and also rewrites the algorithm as a sequence of random continuous operators in the primal-dual space. We compare our analysis with a similar argument by Alacaoglu et al., and give sufficient conditions for an unproven claim in their proof.

[1]  Guanghui Lan,et al.  Randomized First-Order Methods for Saddle Point Optimization , 2014, 1409.8625.

[2]  Patrick L. Combettes,et al.  Stochastic Quasi-Fejér Block-Coordinate Fixed Point Iterations with Random Sweeping , 2014 .

[3]  H. Robbins,et al.  A Convergence Theorem for Non Negative Almost Supermartingales and Some Applications , 1985 .

[4]  F ChanTony,et al.  A General Framework for a Class of First Order Primal-Dual Algorithms for Convex Optimization in Imaging Science , 2010 .

[5]  Ion Necoara,et al.  Nonasymptotic convergence of stochastic proximal point methods for constrained convex optimization , 2017, J. Mach. Learn. Res..

[6]  On the convergence of stochastic primal-dual hybrid gradient , 2019, 1911.00799.

[7]  Volkan Cevher,et al.  Convex Optimization for Big Data: Scalable, randomized, and parallel algorithms for big data analytics , 2014, IEEE Signal Processing Magazine.

[8]  Yuchen Zhang,et al.  Stochastic Primal-Dual Coordinate Method for Regularized Empirical Risk Minimization , 2014, ICML.

[9]  Shai Shalev-Shwartz,et al.  Stochastic dual coordinate ascent methods for regularized loss , 2012, J. Mach. Learn. Res..

[10]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[11]  Yangyang Xu,et al.  Randomized Primal–Dual Proximal Block Coordinate Updates , 2016, Journal of the Operations Research Society of China.

[12]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[13]  Volkan Cevher,et al.  Almost surely constrained convex optimization , 2019, ICML.

[14]  J. Pesquet,et al.  A Class of Randomized Primal-Dual Algorithms for Distributed Optimization , 2014, 1406.6404.

[15]  Heinz H. Bauschke,et al.  Convex Analysis and Monotone Operator Theory in Hilbert Spaces , 2011, CMS Books in Mathematics.

[16]  Antonin Chambolle,et al.  Stochastic Primal-Dual Hybrid Gradient Algorithm with Arbitrary Sampling and Imaging Applications , 2017, SIAM J. Optim..

[17]  Antonin Chambolle,et al.  A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging , 2011, Journal of Mathematical Imaging and Vision.

[18]  Antonin Chambolle,et al.  An introduction to continuous optimization for imaging , 2016, Acta Numerica.

[19]  Tony F. Chan,et al.  A General Framework for a Class of First Order Primal-Dual Algorithms for Convex Optimization in Imaging Science , 2010, SIAM J. Imaging Sci..

[20]  Pascal Bianchi,et al.  A Coordinate-Descent Primal-Dual Algorithm with Large Step Size and Possibly Nonseparable Functions , 2015, SIAM J. Optim..

[21]  Panagiotis Patrinos,et al.  A New Randomized Block-Coordinate Primal-Dual Proximal Algorithm for Distributed Optimization , 2017, IEEE Transactions on Automatic Control.

[22]  Daniel Cremers,et al.  An algorithm for minimizing the Mumford-Shah functional , 2009, 2009 IEEE 12th International Conference on Computer Vision.