Single Domain Generalization with Adversarial Memory
This addresses the challenge of domain generalization in data-constrained scenarios, providing a more realistic solution for applications with limited training diversity.
The paper tackles the problem of Single Domain Generalization (SDG), where models must generalize to unseen domains using only a single training domain, by proposing a method that uses an adversarial memory bank for feature augmentation to align training and testing domains in an invariant subspace. It achieves state-of-the-art performance on standard SDG benchmarks.
Domain Generalization (DG) aims to train models that can generalize to unseen testing domains by leveraging data from multiple training domains. However, traditional DG methods rely on the availability of multiple diverse training domains, limiting their applicability in data-constrained scenarios. Single Domain Generalization (SDG) addresses the more realistic and challenging setting by restricting the training data to a single domain distribution. The main challenges in SDG stem from the limited diversity of training data and the inaccessibility of unseen testing data distributions. To tackle these challenges, we propose a single domain generalization method that leverages an adversarial memory bank to augment training features. Our memory-based feature augmentation network maps both training and testing features into an invariant subspace spanned by diverse memory features, implicitly aligning the training and testing domains in the projected space. To maintain a diverse and representative feature memory bank, we introduce an adversarial feature generation method that creates features extending beyond the training domain distribution. Experimental results demonstrate that our approach achieves state-of-the-art performance on standard single domain generalization benchmarks.