Optimal Task Order for Continual Learning of Multiple Tasks

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Li, Ziyan, Hiratani, Naoki
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908456466776064
author Li, Ziyan
Hiratani, Naoki
author_facet Li, Ziyan
Hiratani, Naoki
contents Continual learning of multiple tasks remains a major challenge for neural networks. Here, we investigate how task order influences continual learning and propose a strategy for optimizing it. Leveraging a linear teacher-student model with latent factors, we derive an analytical expression relating task similarity and ordering to learning performance. Our analysis reveals two principles that hold under a wide parameter range: (1) tasks should be arranged from the least representative to the most typical, and (2) adjacent tasks should be dissimilar. We validate these rules on both synthetic data and real-world image classification datasets (Fashion-MNIST, CIFAR-10, CIFAR-100), demonstrating consistent performance improvements in both multilayer perceptrons and convolutional neural networks. Our work thus presents a generalizable framework for task-order optimization in task-incremental continual learning.
format Preprint
id arxiv_https___arxiv_org_abs_2502_03350
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optimal Task Order for Continual Learning of Multiple Tasks
Li, Ziyan
Hiratani, Naoki
Machine Learning
Continual learning of multiple tasks remains a major challenge for neural networks. Here, we investigate how task order influences continual learning and propose a strategy for optimizing it. Leveraging a linear teacher-student model with latent factors, we derive an analytical expression relating task similarity and ordering to learning performance. Our analysis reveals two principles that hold under a wide parameter range: (1) tasks should be arranged from the least representative to the most typical, and (2) adjacent tasks should be dissimilar. We validate these rules on both synthetic data and real-world image classification datasets (Fashion-MNIST, CIFAR-10, CIFAR-100), demonstrating consistent performance improvements in both multilayer perceptrons and convolutional neural networks. Our work thus presents a generalizable framework for task-order optimization in task-incremental continual learning.
title Optimal Task Order for Continual Learning of Multiple Tasks
topic Machine Learning
url https://arxiv.org/abs/2502.03350