MISA: Mutual Information-Driven Separator With Spectral Alignment
Year: 2026
Type: article
Abstract: With the continual advancement of computer vision techniques, the efficacy of deep learning models has markedly improved. Nonetheless, these accomplishments depend on the premise that both training and test data adhere to the independent and identically distributed (i.i.d.) property. This assumption is frequently contravened in practice due to the variability of unobserved data distributions, resulting in performance deterioration. These variations are named domains, and this has motivated the study of robustness to domain shifts, commonly referred to as domain generalization (DG). In this context, DG methodologies have emerged as a novel research paradigm. This research introduces a domain generalization framework, MISA (Mutual Information-driven Separator with Spectral Alignment), which disentangles and learns semantic features via mutual information optimization and spectral alignment. MISA incorporates an attention mechanism and a domain classifier to extract domain information, while three complementing loss functions directly promote feature disentanglement. Additionally, we conceptually establish that the suggested components diminish generalization error in unseen domains and empirically verify the efficacy of MISA by comprehensive comparisons with existing methodologies.
Source: IEEE Access
Authors: Sihwa Lee, Taehun Lee, Yoon-Yeong Kim
Institution: University of Seoul
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,075
OpenAlex ID: W7170552257
