Mixstyle-Entropy: Whole Process Domain Generalization with Causal Intervention and Perturbation


Luyao Tang (Xiamen University), Yuxuan Yuan (Xiamen University), Chaoqi Chen (The University of Hong Kong), Xinghao Ding (Xiamen University), Yue Huang (Xiamen University)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)Supplementary

Abstract

Despite the considerable advancements achieved by deep neural networks, their performance tends to degenerate when the test environment diverges from the training ones. Domain generalization (DG) solves this issue by learning representations independent of domain-related information, thus facilitating extrapolation to unseen environments. Existing approaches typically focus on formulating tailored training objectives to extract shared features from the source data. However, the disjointed training and testing procedures may compromise robustness, particularly in the face of unforeseen variations during deployment. In this paper, we propose a novel and holistic framework based on causality, named Mixstyle-Entropy, designed to enhance model generalization by incorporating causal intervention during training and causal perturbation during testing. Specifically, during the training phase, we employ entropy-based causal intervention (EnIn) to refine the selection of causal variables. To identify samples with anti-interference causal variables from the target domain, we propose a novel metric, homeostatic score, through causal perturbation (HoPer) to construct a prototype classifier in test time. Experimental results across multiple cross-domain tasks confirm the efficacy of Mixstyle-Entropy.

Citation

@inproceedings{Tang_2024_BMVC,
author    = {Luyao Tang and Yuxuan Yuan and Chaoqi Chen and Xinghao Ding and Yue Huang},
title     = {Mixstyle-Entropy: Whole Process Domain Generalization with Causal Intervention and Perturbation},
booktitle = {35th British Machine Vision Conference 2024, {BMVC} 2024, Glasgow, UK, November 25-28, 2024},
publisher = {BMVA},
year      = {2024},
url       = {https://papers.bmvc2024.org/0212.pdf}
}


Copyright © 2024 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection