Learning domain-invariant features through channel-level sparsification for Out-Of Distribution Generalization
This work addresses the challenge of inconsistent performance across data sources for image analysis systems, representing an incremental improvement over current invariance learning techniques.
The paper tackles the problem of Out-of-Distribution (OOD) generalization in image analysis by addressing shortcut dependencies on domain-specific features, proposing Hierarchical Causal Dropout (HCD) to separate causal from spurious features, resulting in better performance than existing top-tier methods on OOD benchmarks.
Out-of-Distribution (OOD) generalization has become a primary metric for evaluating image analysis systems. Since deep learning models tend to capture domain-specific context, they often develop shortcut dependencies on these non-causal features, leading to inconsistent performance across different data sources. Current techniques, such as invariance learning, attempt to mitigate this. However, they struggle to isolate highly mixed features within deep latent spaces. This limitation prevents them from fully resolving the shortcut learning problem.In this paper, we propose Hierarchical Causal Dropout (HCD), a method that uses channel-level causal masks to enforce feature sparsity. This approach allows the model to separate causal features from spurious ones, effectively performing a causal intervention at the representation level. The training is guided by a Matrix-based Mutual Information (MMI) objective to minimize the mutual information between latent features and domain labels, while simultaneously maximizing the information shared with class labels.To ensure stability, we incorporate a StyleMix-driven VICReg module, which prevents the masks from accidentally filtering out essential causal data. Experimental results on OOD benchmarks show that HCD performs better than existing top-tier methods.