Causal Inference via Style Transfer for Out-of-distribution Generalisation
Open Access
- 4 August 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1746-1757
Abstract
Out-of-distribution (OOD) generalisation aims to build a model that can generalise well on an unseen target domain using knowledge from multiple source domains. To this end, the model should seek the causal dependence between inputs and labels, which may be determined by the semantics of inputs and remain invariant across domains. However, statistical or non-causal methods often cannot capture this dependence and perform poorly due to not considering spurious correlations learnt from model training via unobserved confounders. A well-known existing causal inference method like back-door adjustment cannot be applied to remove spurious correlations as it requires the observation of confounders. In this paper, we propose a novel method that effectively deals with hidden confounders by successfully implementing front-door adjustment (FA). FA requires the choice of a mediator, which we regard as the semantic information of images that helps access the causal mechanism without the need for observing confounders. Further, we propose to estimate the combination of the mediator with other observed images in the front-door formula via style transfer algorithms. Our use of style transfer to estimate FA is novel and sensible for OOD generalisation, which we justify by extensive experimental results on widely used benchmark datasets.Keywords
All Related Versions
This publication has 26 references indexed in Scilit:
- Deep Domain Generalization via Conditional Invariant Adversarial Networks2018
- Identifying causal effects with proxy variables of an unmeasured confounderBiometrika, 2018
- Unified Deep Supervised Domain Adaptation and Generalization2017
- Deeper, Broader and Artier Domain Generalization2017
- Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization2017
- Deep Hashing Network for Unsupervised Domain Adaptation2017
- ImageNet: A large-scale hierarchical image database2009
- The Nature of Statistical Learning Theory2000
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Causal diagrams for empirical researchBiometrika, 1995