Saved in:
Bibliographic Details
Main Authors: Yang, Honglin, Ma, Ji, Yu, Xiao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.12030
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913614380662784
author Yang, Honglin
Ma, Ji
Yu, Xiao
author_facet Yang, Honglin
Ma, Ji
Yu, Xiao
contents The optimization-based meta-learning approach is gaining increased traction because of its unique ability to quickly adapt to a new task using only small amounts of data. However, existing optimization-based meta-learning approaches, such as MAML, ANIL and their variants, generally employ backpropagation for upper-level gradient estimation, which requires using historical lower-level parameters/gradients and thus increases computational and memory overhead in each iteration. In this paper, we propose a meta-learning algorithm that can avoid using historical parameters/gradients and significantly reduce memory costs in each iteration compared to existing optimization-based meta-learning approaches. In addition to memory reduction, we prove that our proposed algorithm converges sublinearly with the iteration number of upper-level optimization, and the convergence error decays sublinearly with the batch size of sampled tasks. In the specific case in terms of deterministic meta-learning, we also prove that our proposed algorithm converges to an exact solution. Moreover, we quantify that the computational complexity of the algorithm is on the order of $\mathcal{O}(ε^{-1})$, which matches existing convergence results on meta-learning even without using any historical parameters/gradients. Experimental results on meta-learning benchmarks confirm the efficacy of our proposed algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2412_12030
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Memory-Reduced Meta-Learning with Guaranteed Convergence
Yang, Honglin
Ma, Ji
Yu, Xiao
Machine Learning
Optimization and Control
The optimization-based meta-learning approach is gaining increased traction because of its unique ability to quickly adapt to a new task using only small amounts of data. However, existing optimization-based meta-learning approaches, such as MAML, ANIL and their variants, generally employ backpropagation for upper-level gradient estimation, which requires using historical lower-level parameters/gradients and thus increases computational and memory overhead in each iteration. In this paper, we propose a meta-learning algorithm that can avoid using historical parameters/gradients and significantly reduce memory costs in each iteration compared to existing optimization-based meta-learning approaches. In addition to memory reduction, we prove that our proposed algorithm converges sublinearly with the iteration number of upper-level optimization, and the convergence error decays sublinearly with the batch size of sampled tasks. In the specific case in terms of deterministic meta-learning, we also prove that our proposed algorithm converges to an exact solution. Moreover, we quantify that the computational complexity of the algorithm is on the order of $\mathcal{O}(ε^{-1})$, which matches existing convergence results on meta-learning even without using any historical parameters/gradients. Experimental results on meta-learning benchmarks confirm the efficacy of our proposed algorithm.
title Memory-Reduced Meta-Learning with Guaranteed Convergence
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2412.12030