CycleMix: Mixing Source Domains for Domain Generalization in Style-Dependent Data
This addresses the issue of spurious style-class correlations in training data for machine learning systems, though it is incremental as it builds on existing CycleGAN and domain generalization methods.
The paper tackled the domain generalization problem in image classification by training a robust feature extractor to disregard style-dependent features and use style-invariant representations, achieving validation on the PACS benchmark.
As deep learning-based systems have become an integral part of everyday life, limitations in their generalization ability have begun to emerge. Machine learning algorithms typically rely on the i.i.d. assumption, meaning that their training and validation data are expected to follow the same distribution, which does not necessarily hold in practice. In the case of image classification, one frequent reason that algorithms fail to generalize is that they rely on spurious correlations present in training data, such as associating image styles with target classes. These associations may not be present in the unseen test data, leading to significant degradation of their effectiveness. In this work, we attempt to mitigate this Domain Generalization (DG) problem by training a robust feature extractor which disregards features attributed to image-style but infers based on style-invariant image representations. To achieve this, we train CycleGAN models to learn the different styles present in the training data and randomly mix them together to create samples with novel style attributes to improve generalization. Experimental results on the PACS DG benchmark validate the proposed method.