Riemannian Bilevel Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dutta, Sanchayan, Cheng, Xiang, Sra, Suvrit
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929357967065088
author Dutta, Sanchayan
Cheng, Xiang
Sra, Suvrit
author_facet Dutta, Sanchayan
Cheng, Xiang
Sra, Suvrit
contents We develop new algorithms for Riemannian bilevel optimization. We focus in particular on batch and stochastic gradient-based methods, with the explicit goal of avoiding second-order information such as Riemannian hyper-gradients. We propose and analyze $\mathrm{RF^2SA}$, a method that leverages first-order gradient information to navigate the complex geometry of Riemannian manifolds efficiently. Notably, $\mathrm{RF^2SA}$ is a single-loop algorithm, and thus easier to implement and use. Under various setups, including stochastic optimization, we provide explicit convergence rates for reaching $ε$-stationary points. We also address the challenge of optimizing over Riemannian manifolds with constraints by adjusting the multiplier in the Lagrangian, ensuring convergence to the desired solution without requiring access to second-order derivatives.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15816
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Riemannian Bilevel Optimization
Dutta, Sanchayan
Cheng, Xiang
Sra, Suvrit
Optimization and Control
Artificial Intelligence
Machine Learning
We develop new algorithms for Riemannian bilevel optimization. We focus in particular on batch and stochastic gradient-based methods, with the explicit goal of avoiding second-order information such as Riemannian hyper-gradients. We propose and analyze $\mathrm{RF^2SA}$, a method that leverages first-order gradient information to navigate the complex geometry of Riemannian manifolds efficiently. Notably, $\mathrm{RF^2SA}$ is a single-loop algorithm, and thus easier to implement and use. Under various setups, including stochastic optimization, we provide explicit convergence rates for reaching $ε$-stationary points. We also address the challenge of optimizing over Riemannian manifolds with constraints by adjusting the multiplier in the Lagrangian, ensuring convergence to the desired solution without requiring access to second-order derivatives.
title Riemannian Bilevel Optimization
topic Optimization and Control
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2405.15816