Riemannian Bilevel Optimization with Gradient Aggregation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Zhuo, Xu, Xinjian, Ying, Shihui, Zeng, Tieyong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908599126589440
author Chen, Zhuo
Xu, Xinjian
Ying, Shihui
Zeng, Tieyong
author_facet Chen, Zhuo
Xu, Xinjian
Ying, Shihui
Zeng, Tieyong
contents Bilevel optimization (BLO) offers a principled framework for hierarchical decision-making and has been widely applied in machine learning tasks such as hyperparameter optimization and meta-learning. While existing BLO methods are mostly developed in Euclidean spaces, many real-world problems involve structural constraints. In this paper, we propose a Riemannian bilevel optimization (RBLO) algorithm that incorporates a bilevel descent aggregation (BDA) scheme to jointly coordinate upper- and lower-level updates. Concretely, first we abstract the constraints in the BLO to a manifold structure and then transform the constrained BLO be a unconstrained RBLO problem. Second, to address limitations of existing RBLO methods, particularly the restrictive assumptions required for convergence, we reformulate the bilevel problem using smooth manifold mappings and provide a convergence analysis under the conditions of geodesic convexity and Lipschitz smoothness. Finally, we recall the multi-view hypergraph spectral clustering task, and evaluate the proposed approach on 3sources data sets. The numerical results validate the superior performance over Euclidean and manifold-based baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15305
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Riemannian Bilevel Optimization with Gradient Aggregation
Chen, Zhuo
Xu, Xinjian
Ying, Shihui
Zeng, Tieyong
Optimization and Control
Numerical Analysis
49Q12, 65K10
Bilevel optimization (BLO) offers a principled framework for hierarchical decision-making and has been widely applied in machine learning tasks such as hyperparameter optimization and meta-learning. While existing BLO methods are mostly developed in Euclidean spaces, many real-world problems involve structural constraints. In this paper, we propose a Riemannian bilevel optimization (RBLO) algorithm that incorporates a bilevel descent aggregation (BDA) scheme to jointly coordinate upper- and lower-level updates. Concretely, first we abstract the constraints in the BLO to a manifold structure and then transform the constrained BLO be a unconstrained RBLO problem. Second, to address limitations of existing RBLO methods, particularly the restrictive assumptions required for convergence, we reformulate the bilevel problem using smooth manifold mappings and provide a convergence analysis under the conditions of geodesic convexity and Lipschitz smoothness. Finally, we recall the multi-view hypergraph spectral clustering task, and evaluate the proposed approach on 3sources data sets. The numerical results validate the superior performance over Euclidean and manifold-based baselines.
title Riemannian Bilevel Optimization with Gradient Aggregation
topic Optimization and Control
Numerical Analysis
49Q12, 65K10
url https://arxiv.org/abs/2510.15305