Task-Level Self-Supervision for Cross-Domain Few-Shot Learning

Authors

  • Wang Yuan East China Normal University
  • Zhizhong Zhang East China Normal University
  • Cong Wang Huawei Technologies
  • Haichuan Song East China Normal University
  • Yuan Xie East China Normal University
  • Lizhuang Ma East China Normal University Shanghai Jiao Tong University MoE Key Lab of Artificial Intelligence, Shanghai Jiao Tong University

DOI:

https://doi.org/10.1609/aaai.v36i3.20230

Keywords:

Computer Vision (CV)

Abstract

Learning with limited labeled data is a long-standing problem. Among various solutions, episodic training progres-sively classifies a series of few-shot tasks and thereby is as-sumed to be beneficial for improving the model’s generalization ability. However, recent studies show that it is eveninferior to the baseline model when facing domain shift between base and novel classes. To tackle this problem, we pro-pose a domain-independent task-level self-supervised (TL-SS) method for cross-domain few-shot learning.TL-SS strategy promotes the general idea of label-based instance-levelsupervision to task-level self-supervision by augmenting mul-tiple views of tasks. Two regularizations on task consistencyand correlation metric are introduced to remarkably stabi-lize the training process and endow the generalization ability into the prediction model. We also propose a high-order associated encoder (HAE) being adaptive to various tasks.By utilizing 3D convolution module, HAE is able to generate proper parameters and enables the encoder to flexibly toany unseen tasks. Two modules complement each other andshow great promotion against state-of-the-art methods experimentally. Finally, we design a generalized task-agnostic test,where our intriguing findings highlight the need to re-think the generalization ability of existing few-shot approaches.

Downloads

Published

2022-06-28

How to Cite

Yuan, W., Zhang, Z., Wang, C., Song, H., Xie, Y., & Ma, L. (2022). Task-Level Self-Supervision for Cross-Domain Few-Shot Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 36(3), 3215-3223. https://doi.org/10.1609/aaai.v36i3.20230

Issue

Section

AAAI Technical Track on Computer Vision III