Out-of-Domain Generalization from a Single Source: An Uncertainty Quantification Approach
This addresses a worst-case scenario for machine learning models that need to generalize to unseen domains with only one source available, representing an incremental improvement in domain generalization methods.
The paper tackles the problem of out-of-domain generalization from a single training domain by proposing Meta-Learning based Adversarial Domain Augmentation, which uses adversarial training and uncertainty quantification to create challenging fictitious populations, achieving superior performance on benchmark datasets.
We are concerned with a worst-case scenario in model generalization, in the sense that a model aims to perform well on many unseen domains while there is only one single domain available for training. We propose Meta-Learning based Adversarial Domain Augmentation to solve this Out-of-Domain generalization problem. The key idea is to leverage adversarial training to create "fictitious" yet "challenging" populations, from which a model can learn to generalize with theoretical guarantees. To facilitate fast and desirable domain augmentation, we cast the model training in a meta-learning scheme and use a Wasserstein Auto-Encoder to relax the widely used worst-case constraint. We further improve our method by integrating uncertainty quantification for efficient domain generalization. Extensive experiments on multiple benchmark datasets indicate its superior performance in tackling single domain generalization.