Improving the Generalization of Meta-learning on Unseen Domains via Adversarial Shift

Abstract

Meta-learning provides a promising way for learning to efficiently learn and achieves great success in many applications. However, most meta-learning literature focuses on dealing with tasks from a same domain, making it brittle to generalize to tasks from the other unseen domains. In this work, we address this problem by simulating tasks from the other unseen domains to improve the generalization and robustness of meta-learning method. Specifically, we propose a model-agnostic shift layer to learn how to simulate the domain shift and generate pseudo tasks, and develop a new adversarial learning-to-learn mechanism to train it. Based on the pseudo tasks, the meta-learning model can learn cross-domain meta-knowledge, which can generalize well on unseen domains. We conduct extensive experiments under the domain generalization setting. Experimental results demonstrate that the proposed shift layer is applicable to various meta-learning frameworks. Moreover, our method also leads to state-of-the-art performance on different cross-domain few-shot classification benchmarks and produces good results on cross-domain few-shot regression.

arXiv commentv1

References in corpus (12)

• Prototypical Networks for Few-shot Learning
• Deep Visual Domain Adaptation: A Survey
• A Learned Representation For Artistic Style
• A Survey on Multi-Task Learning
• Certifying Some Distributional Robustness with Principled Adversarial Training
• Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables
• A Closer Look at Few-shot Classification
• Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation
• Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
• Cross-Domain Few-Shot Learning by Representation Fusion
• Explanation-Guided Training for Cross-Domain Few-Shot Classification
• Domain Generalization via Inference-time Label-Preserving Target Projections