Trilevel and Multilevel Optimization using Monotone Operator Theory

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Shafiei, Allahkaram, Kungurtsev, Vyacheslav, Marecek, Jakub
Natura: Preprint
Pubblicazione: 2021
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916224962658304
author Shafiei, Allahkaram
Kungurtsev, Vyacheslav
Marecek, Jakub
author_facet Shafiei, Allahkaram
Kungurtsev, Vyacheslav
Marecek, Jakub
contents We consider rather a general class of multi-level optimization problems, where a convex objective function is to be minimized subject to constraints of optimality of nested convex optimization problems. As a special case, we consider a trilevel optimization problem, where the objective of the two lower layers consists of a sum of a smooth and a non-smooth term.~Based on fixed-point theory and related arguments, we present a natural first-order algorithm and analyze its convergence and rates of convergence in several regimes of parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2105_09407
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Trilevel and Multilevel Optimization using Monotone Operator Theory
Shafiei, Allahkaram
Kungurtsev, Vyacheslav
Marecek, Jakub
Optimization and Control
Artificial Intelligence
We consider rather a general class of multi-level optimization problems, where a convex objective function is to be minimized subject to constraints of optimality of nested convex optimization problems. As a special case, we consider a trilevel optimization problem, where the objective of the two lower layers consists of a sum of a smooth and a non-smooth term.~Based on fixed-point theory and related arguments, we present a natural first-order algorithm and analyze its convergence and rates of convergence in several regimes of parameters.
title Trilevel and Multilevel Optimization using Monotone Operator Theory
topic Optimization and Control
Artificial Intelligence
url https://arxiv.org/abs/2105.09407