A Nested Bi-level Optimization Framework for Robust Few Shot Learning

TL;DR

This paper introduces NestedMAML, a robust meta-learning framework that assigns adaptive weights to training tasks or instances via nested bi-level optimization, improving performance in domain-shifted and noisy data scenarios.

Contribution

NestedMAML is a novel algorithm that optimizes task weights through nested bi-level optimization, enhancing robustness in few-shot learning under domain shift and noisy data.

Findings

Outperforms state-of-the-art robust meta-learning methods.

Effectively mitigates effects of noisy labels and domain shift.

Demonstrates significant improvements on synthetic and real-world datasets.

Abstract

Model-Agnostic Meta-Learning (MAML), a popular gradient-based meta-learning framework, assumes that the contribution of each task or instance to the meta-learner is equal. Hence, it fails to address the domain shift between base and novel classes in few-shot learning. In this work, we propose a novel robust meta-learning algorithm, NestedMAML, which learns to assign weights to training tasks or instances. We consider weights as hyper-parameters and iteratively optimize them using a small set of validation tasks set in a nested bi-level optimization approach (in contrast to the standard bi-level optimization in MAML). We then apply NestedMAML in the meta-training stage, which involves (1) several tasks sampled from a distribution different from the meta-test task distribution, or (2) some data samples with noisy labels. Extensive experiments on synthetic and real-world datasets demonstrate that NestedMAML efficiently mitigates the effects of "unwanted" tasks or instances, leading to significant improvement over the state-of-the-art robust meta-learning methods.