Arrow Research search
Back to NeurIPS

NeurIPS 2021

Online Meta-Learning via Learning with Layer-Distributed Memory

Conference Paper Artificial Intelligence ยท Machine Learning

Abstract

We demonstrate that efficient meta-learning can be achieved via end-to-end training of deep neural networks with memory distributed across layers. The persistent state of this memory assumes the entire burden of guiding task adaptation. Moreover, its distributed nature is instrumental in orchestrating adaptation. Ablation experiments demonstrate that providing relevant feedback to memory units distributed across the depth of the network enables them to guide adaptation throughout the entire network. Our results show that this is a successful strategy for simplifying meta-learning -- often cast as a bi-level optimization problem -- to standard end-to-end training, while outperforming gradient-based, prototype-based, and other memory-based meta-learning strategies. Additionally, our adaptation strategy naturally handles online learning scenarios with a significant delay between observing a sample and its corresponding label -- a setting in which other approaches struggle. Adaptation via distributed memory is effective across a wide range of learning tasks, ranging from classification to online few-shot semantic segmentation.

Authors

Keywords

No keywords are indexed for this paper.

Context

Venue
Annual Conference on Neural Information Processing Systems
Archive span
1987-2025
Indexed papers
30776
Paper id
1021404798422452703
v2026.09.13