Cross-domain few-shot learning with unlabelled data

2021/01/19 by Yao, Fupin · 1 citation
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2101.07899

Abstract

Few shot learning aims to solve the data scarcity problem. If there is a domain shift between the test set and the training set, their performance will decrease a lot. This setting is called Cross-domain few-shot learning. However, this is very challenging because the target domain is unseen during training. Thus we propose a new setting some unlabelled data from the target domain is provided, which can bridge the gap between the source domain and the target domain. A benchmark for this setting is constructed using DomainNet \citepeng2018oment. We come up with a self-supervised learning method to fully utilize the knowledge in the labeled training set and the unlabelled set. Extensive experiments show that our methods outperforms several baseline methods by a large margin. We also carefully design an episodic training pipeline which yields a significant performance boost.

Cited by

• FreqGRL: Suppressing Low-Frequency Bias and Mining High-Frequency Knowledge for Cross-Domain Few-Shot Learning

Related

• Understanding Cross-Domain Few-Shot Learning Based on Domain Similarity and Few-Shot Difficulty
• Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data
• Self-training for Few-shot Transfer Across Extreme Task Differences
• A Cross-Domain Few-Shot Learning Method Based on Domain Knowledge Mapping
• Cross-Domain Cross-Set Few-Shot Learning via Learning Compact and Aligned Representations