A Zeroth-Order Proximal Stochastic Gradient Method for Weakly Convex Stochastic Optimization

DOI10.1137/22m1494270arXiv2205.01633MaRDI QIDQ6066421

Spyridon Pougkakiotis, Unnamed Author

Publication date: 16 November 2023

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2205.01633

zbMATH Keywords

stochastic gradient descentcomposite optimizationhyperparameter tuningzeroth-order optimizationweakly convex stochastic optimization

Mathematics Subject Classification ID

Nonlinear programming (90C30) Derivative-free methods and methods using generalized derivatives (90C56) Stochastic programming (90C15)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
• Generalized ADMM with optimal indefinite proximal term for linearly constrained convex optimization
• Residual whiteness principle for automatic parameter selection in \(\ell_2-\ell_2\) image super-resolution problems
• Convergence of a random optimization method for constrained optimization problems
• GCV for Tikhonov regularization by partial SVD
• An efficient duality-based approach for PDE-constrained sparse optimization
• \(L\)-curve curvature bounds via Lanczos bidiagonalization
• A zeroth order method for stochastic weakly convex optimization
• Noisy zeroth-order optimization for non-smooth saddle point problems
• Efficiency of minimizing compositions of convex functions and smooth maps
• Random gradient-free minimization of convex functions
• On the global and linear convergence of the generalized alternating direction method of multipliers
• Phase retrieval: stability and recovery guarantees
• Random optimization
• Zeroth-order nonconvex stochastic optimization: handling constraints, high dimensionality, and saddle points
• Optimal Rates for Zero-Order Convex Optimization: The Power of Two Function Evaluations
• Strong and Weak Convexity of Sets and Functions
• Convergence and regularization results for optimal control problems with sparsity functional
• Self-calibration and biconvex compressive sensing
• Algorithm 866
• Introduction to Derivative-Free Optimization
• Minimization by Random Search Techniques
• Multivariate stochastic approximation using a simultaneous perturbation gradient approximation
• Generalized Gradients and Applications
• Variational Analysis
• Stochastic Model-Based Minimization of Weakly Convex Functions
• Pattern Search Methods for User-Provided Points: Application to Molecular Geometry Problems
• Zeroth-Order Stochastic Compositional Algorithms for Risk-Aware Learning
• IFISS: A Computational Laboratory for Investigating Incompressible Flow Problems
• Interior‐point methods and preconditioning for PDE‐constrained optimization problems involving sparsity terms
• Optimization by Direct Search in Matrix Computations
• Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic Programming
• Proximité et dualité dans un espace hilbertien
• Finding Optimal Algorithmic Parameters Using Derivative‐Free Optimization
• Zeroth-order optimization with orthogonal random directions

This page was built for publication: A Zeroth-Order Proximal Stochastic Gradient Method for Weakly Convex Stochastic Optimization