Domain Generalization via Multidomain Discriminant Analysis
This addresses the problem of domain shift in machine learning for practitioners needing robust models, but it is incremental as it builds on existing domain generalization methods.
The paper tackles domain generalization for classification by proposing Multidomain Discriminant Analysis (MDA), which learns a domain-invariant feature transformation to improve generalization on unseen target domains, with experiments showing its effectiveness on synthetic and real benchmark datasets.
Domain generalization (DG) aims to incorporate knowledge from multiple source domains into a single model that could generalize well on unseen target domains. This problem is ubiquitous in practice since the distributions of the target data may rarely be identical to those of the source data. In this paper, we propose Multidomain Discriminant Analysis (MDA) to address DG of classification tasks in general situations. MDA learns a domain-invariant feature transformation that aims to achieve appealing properties, including a minimal divergence among domains within each class, a maximal separability among classes, and overall maximal compactness of all classes. Furthermore, we provide the bounds on excess risk and generalization error by learning theory analysis. Comprehensive experiments on synthetic and real benchmark datasets demonstrate the effectiveness of MDA.