Robust Domain Generalization under Divergent Marginal and Conditional Distributions
Authors:
- Jewon YeomJYJewon Yeom
- Kyubyung ChaeKCKyubyung Chae
- Hyunggyu LimHLHyunggyu Lim
- Yoonna OhYOYoonna Oh
- Dongyoon YangDYDongyoon Yang
- Taesup KimTKTaesup Kim
Preprint
Abstract
Domain generalization (DG) aims to learn predictive models that can generalize to unseen domains. Most existing DG approaches focus on learning domain-invariant representations under the assumption of conditional distribution shift (i.e., primarily addressing changes in $P(X\mid Y)$ while assuming $P(Y)$ remains stable). However, real-world scenarios with multiple domains often involve compound distribution shifts where both the marginal label distribution $P(Y)$ and the conditional distribution $P(X\mid Y)$ vary simultaneously. To address this, we propose a unified framework for robust domain generalization under divergent marginal and conditional distributions. We derive a novel risk bound for unseen domains by explicitly decomposing the joint distribution into marginal and conditional components and characterizing risk gaps arising from both sources of divergence. To operationalize this bound, we design a meta-learning procedure that minimizes and validates the proposed risk bound across seen domains, ensuring strong generalization to unseen ones. Empirical evaluations demonstrate that our method achieves state-of-the-art performance not only on conventional DG benchmarks but also in challenging multi-domain long-tailed recognition settings where both marginal and conditional shifts are pronounced.
Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- Balanced Direction from Multifarious Choices: Arithmetic Meta-Learning for Domain Generalization2025
- Multi-Domain Long-Tailed Learning: Challenges, Progress, and ProspectsIEEE Access, 2024
- Meta-learning the invariant representation for domain generalizationMachine Learning, 2022
- On Multi-Domain Long-Tailed Recognition, Imbalanced Domain Generalization and Beyond2022
- Deep imbalanced domain adaptation for transfer learning fault diagnosis of bearings under multiple working conditionsReliability Engineering & System Safety, 2022
- Causality Inspired Representation Learning for Domain Generalization2022
- Loss landscapes and optimization in over-parameterized non-linear systems and neural networksApplied and Computational Harmonic Analysis, 2022
- Deep Domain Generalization via Conditional Invariant Adversarial Networks2018
- A systematic study of the class imbalance problem in convolutional neural networksNeural Networks, 2018
- ImageNet: A large-scale hierarchical image database2009