Most few-shot learning techniques are pre-trained on a large, labeled "base dataset". In problem domains where such large labeled datasets are not available for pre-training (e.g., X-ray, satellite images), one must resort to pre-training in a different "source" problem domain (e.g., ImageNet), which can be very different from the desired target task. Traditional few-shot and transfer learning techniques fail in the presence of such extreme differences between the source and target tasks. In this paper, we present a simple and effective solution to tackle this extreme domain gap: self-training a source domain representation on unlabeled data from the target domain. We show that this improves one-shot performance on the target domain by 2.9 points on average on the challenging BSCD-FSL benchmark consisting of datasets from multiple domains. Our code is available at https://github.com/cpphoo/STARTUP.
@article{arxiv.2010.07734,
title = {Self-training for Few-shot Transfer Across Extreme Task Differences},
author = {Cheng Perng Phoo and Bharath Hariharan},
journal= {arXiv preprint arXiv:2010.07734},
year = {2021}
}
Comments
Published as a conference paper at ICLR 2021(oral)