DOMAIN GENERALIZED FEW-SHOT IMAGE CLASSIFICATION VIA META REGULARIZATION NETWORK

In few-shot image classification scenarios, meta-learning methods aim to learn transferable feature representations extracted from seen domains (base classes) in the meta-training phase and quickly adapt to unseen domains (novel classes) in the meta-testing phase. However, when seen and unseen domains have a large discrepancy, existing approaches do not perform well due to the incapability of generalizing to unseen domains. In this paper, we investigate the challenging domain generalized few-shot learning problem. We design an Meta Regularization Network (MRN) to learn a domain-invariant discriminative feature space, where a learning to learn update strategy is used to simulate domain shifts caused by seen and unseen domains in the meta-training phase. The simulation trains the model to learn to reorganize the knowledge acquired from seen domains to represent unseen domains. Extensive experiments and analysis show that our proposed MRN can significantly improve the generalization ability of various meta-learning methods to achieve state-of-the-art performance in domain generalized few-shot learning.