Saved in:
Bibliographic Details
Main Authors: Lee, Isaac Ning, Mahmoodi, Leila, Le, Trung, Harandi, Mehrtash
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.18604
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916007753285632
author Lee, Isaac Ning
Mahmoodi, Leila
Le, Trung
Harandi, Mehrtash
author_facet Lee, Isaac Ning
Mahmoodi, Leila
Le, Trung
Harandi, Mehrtash
contents State-Space Models (SSMs) excel at capturing long-range dependencies with structured recurrence, making them well-suited for sequence modeling. However, their evolving internal states pose challenges in adapting them under Continual Learning (CL). This is particularly difficult in exemplar-free settings, where the absence of prior data leaves updates to the dynamic SSM states unconstrained, resulting in catastrophic forgetting. To address this, we propose Inf-SSM, a novel and simple geometry-aware regularization method that utilizes the geometry of the infinite-dimensional Grassmannian to constrain state evolution during CL. Unlike classical continual learning methods that constrain weight updates, Inf-SSM regularizes the infinite-horizon evolution of SSMs encoded in their extended observability subspace. We show that enforcing this regularization requires solving a matrix equation known as the Sylvester equation, which typically incurs $\mathcal{O}(n^3)$ complexity. We develop a $\mathcal{O}(n^2)$ solution by exploiting the structure and properties of SSMs. This leads to an efficient regularization mechanism that can be seamlessly integrated into existing CL methods. Comprehensive experiments on challenging benchmarks, including ImageNet-R and Caltech-256, demonstrate a significant reduction in forgetting while improving accuracy across sequential tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_18604
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exemplar-Free Continual Learning for State Space Models
Lee, Isaac Ning
Mahmoodi, Leila
Le, Trung
Harandi, Mehrtash
Machine Learning
State-Space Models (SSMs) excel at capturing long-range dependencies with structured recurrence, making them well-suited for sequence modeling. However, their evolving internal states pose challenges in adapting them under Continual Learning (CL). This is particularly difficult in exemplar-free settings, where the absence of prior data leaves updates to the dynamic SSM states unconstrained, resulting in catastrophic forgetting. To address this, we propose Inf-SSM, a novel and simple geometry-aware regularization method that utilizes the geometry of the infinite-dimensional Grassmannian to constrain state evolution during CL. Unlike classical continual learning methods that constrain weight updates, Inf-SSM regularizes the infinite-horizon evolution of SSMs encoded in their extended observability subspace. We show that enforcing this regularization requires solving a matrix equation known as the Sylvester equation, which typically incurs $\mathcal{O}(n^3)$ complexity. We develop a $\mathcal{O}(n^2)$ solution by exploiting the structure and properties of SSMs. This leads to an efficient regularization mechanism that can be seamlessly integrated into existing CL methods. Comprehensive experiments on challenging benchmarks, including ImageNet-R and Caltech-256, demonstrate a significant reduction in forgetting while improving accuracy across sequential tasks.
title Exemplar-Free Continual Learning for State Space Models
topic Machine Learning
url https://arxiv.org/abs/2505.18604