DSDRNet: Disentangling Representation and Reconstruct Network for Domain Generalization
This work addresses domain generalization for machine learning models facing distribution shifts, presenting an incremental improvement over existing methods.
The paper tackles domain generalization challenges by proposing DSDRNet, a disentanglement-reconstruction network that integrates inter-instance and intra-instance features with novel supervised signals, and it outperforms other methods on four benchmark datasets.
Domain generalization faces challenges due to the distribution shift between training and testing sets, and the presence of unseen target domains. Common solutions include domain alignment, meta-learning, data augmentation, or ensemble learning, all of which rely on domain labels or domain adversarial techniques. In this paper, we propose a Dual-Stream Separation and Reconstruction Network, dubbed DSDRNet. It is a disentanglement-reconstruction approach that integrates features of both inter-instance and intra-instance through dual-stream fusion. The method introduces novel supervised signals by combining inter-instance semantic distance and intra-instance similarity. Incorporating Adaptive Instance Normalization (AdaIN) into a two-stage cyclic reconstruction process enhances self-disentangled reconstruction signals to facilitate model convergence. Extensive experiments on four benchmark datasets demonstrate that DSDRNet outperforms other popular methods in terms of domain generalization capabilities.