Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Comeau, Jacob, Bazinet, Mathieu, Germain, Pascal, Subakan, Cem
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2503.10503
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918357699133440
author Comeau, Jacob
Bazinet, Mathieu
Germain, Pascal
Subakan, Cem
author_facet Comeau, Jacob
Bazinet, Mathieu
Germain, Pascal
Subakan, Cem
contents Continual learning algorithms aim to learn from a sequence of tasks. In order to avoid catastrophic forgetting, most existing approaches rely on heuristics and do not provide computable learning guarantees. In this paper, we introduce Continual Pick-to-Learn (CoP2L), a method grounded in sample compression theory that retains representative samples for each task in a principled and efficient way. This allows us to derive non-vacuous, numerically computable upper bounds on the generalization loss of the learned predictors after each task. We evaluate CoP2L on standard continual learning benchmarks under Class-Incremental and Task-Incremental settings, showing that it effectively mitigates catastrophic forgetting. It turns out that CoP2L is empirically competitive with baseline methods while certifying predictor reliability in continual learning with a non-vacuous bound.
format Preprint
id arxiv_https___arxiv_org_abs_2503_10503
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sample Compression for Self Certified Continual Learning
Comeau, Jacob
Bazinet, Mathieu
Germain, Pascal
Subakan, Cem
Machine Learning
Continual learning algorithms aim to learn from a sequence of tasks. In order to avoid catastrophic forgetting, most existing approaches rely on heuristics and do not provide computable learning guarantees. In this paper, we introduce Continual Pick-to-Learn (CoP2L), a method grounded in sample compression theory that retains representative samples for each task in a principled and efficient way. This allows us to derive non-vacuous, numerically computable upper bounds on the generalization loss of the learned predictors after each task. We evaluate CoP2L on standard continual learning benchmarks under Class-Incremental and Task-Incremental settings, showing that it effectively mitigates catastrophic forgetting. It turns out that CoP2L is empirically competitive with baseline methods while certifying predictor reliability in continual learning with a non-vacuous bound.
title Sample Compression for Self Certified Continual Learning
topic Machine Learning
url https://arxiv.org/abs/2503.10503