Automatic Visual Instrumental Variable Learning for Confounding-Resistant Domain Generalization

Fuyuan CAO, Shichang Qiao, Kui Yu, Jiye Liang

Keywords: Domain generalization, visual instrumental variable, causal representation learning

TL;DR: We propose a novel approach, VIV-DG, that automatically learns visual instrumental variables to enable confounding-resistant domain generalization, even in the presence of unobserved confounders.

Abstract: Many confounding-resistant domain generalization methods for image classification have been developed based on causal interventions. However, their reliance on strong assumptions limits their effectiveness in handling unobserved confounders. Although recent work introduces instrumental variables (IVs) to overcome this limitation, the reliance on manually predefined instruments, particularly in the context of visual data, may result in severe bias or invalidity when IV conditions are violated. To address these issues, we propose a novel approach to automatically learning Visual Instrumental Variables for confounding-resistant Domain Generalization (VIV-DG). We observe that certain non-causal visual attributes in image data naturally satisfy the basic conditions required for valid IVs. Motivated by this insight, we propose the visual instrumental variable, a novel concept that extends classical IV theory to the visual domain. Furthermore, we develop an automatic visual instrumental variable learner that enforces IV conditions on learned representations, enabling the automatic learning of valid visual instrumental variables from image data. Ultimately, VIV-DG inherits the strengths of classical IVs to mitigate unobserved confounding and avoids the significant bias caused by violations of IV conditions in predefined IVs. Extensive experiments on multiple benchmarks verify that VIV-DG achieves superior generalization ability.

Primary Area: General machine learning (supervised, unsupervised, online, active, etc.)

Submission Number: 18679