paper-with-me

홈 › Papers

Learning where to learn: Gradient sparsity in meta and continual learning

2021-10-27 · NeurIPS 2021 12 · Johannes von Oswald, Dominic Zhao, Seijin Kobayashi, Simon Schug, Massimo Caccia, Nicolas Zucchet, João Sacramento

Finding neural network weights that generalize well from small datasets is difficult. A promising approach is to learn a weight initialization such that a small number of weight changes results in low generalization error. We show that this form of meta-learning can be improved by letting the learning algorithm decide which weights to change, i.e., by learning where to learn. We find that patterned sparsity emerges from this process, with the pattern of sparsity varying on a problem-by-problem basis. This selective sparsity results in better generalization and less interference in a range of few-shot and continual learning problems. Moreover, we find that sparse learning also emerges in a more expressive model where learning rates are meta-learned. Our results shed light on an ongoing debate on whether meta-learning can discover adaptable features and suggest that learning by sparse gradient descent is a powerful inductive bias for meta-learning systems.

📄 PDF Abstract BibTeX arXiv:2110.14402

Code (1)

johswald/learning_where_to_learn 공식 구현 pytorch

Tasks

Continual LearningInductive BiasMeta-LearningSparse Learning

Similar Papers 제목 키워드 기반

Efficient Meta-Learning for Continual Learning with Taylor Expansion Approximation

2022-10-03 · Xiaohan Zou, Tong Lin

Continual learning aims to alleviate catastrophic forgetting when handling consecutive tasks under non-stationary distributions. Gradient-based meta-learning algorithms have shown the capability to implicitly solve the t…

Computational EfficiencyContinual LearningMeta-Learning

Dual-Stage Invariant Continual Learning under Extreme Visual Sparsity

2026-03-27 · Rangya Zhang, Jiaping Xiao, Lu Bai, Yuhang Zhang 외 arxiv

Continual learning seeks to maintain stable adaptation under non-stationary environments, yet this problem becomes particularly challenging in object detection, where most existing methods implicitly assume relatively ba…

Continual LearningObject Detection

Enabling Continual Learning in Neural Networks with Meta Learning

2019-05-16 · ICML Workshop AMTL 2019 6 · Anonymous

Catastrophic forgetting in neural networks is one of the most well-known problems in continual learning. Previous attempts on addressing the problem focus on preventing important weights from changing. Such methods often…

Continual LearningMeta-LearningTransfer Learning

MANGO: Meta-Adaptive Network Gradient Optimization for Online Continual Learning

2026-05-18 · Ankita Awasthi, Marco Apolinario, Kaushik Roy arxiv

In Online Continual Learning (OCL), a neural network sequentially learns from a non-stationary data stream in a single-pass with access only to a limited memory replay buffer. This contrasts sharply with off-line continu…

Class Incremental LearningContinual Learning

Meta Continual Learning via Dynamic Programming

2020-08-05 · R. Krishnan, Prasanna Balaprakash

Meta continual learning algorithms seek to train a model when faced with similar tasks observed in a sequential manner. Despite promising methodological advancements, there is a lack of theoretical frameworks that enable…

Continual Learning