LGMay 10, 2024

Non-stationary Domain Generalization: Theory and Algorithm

arXiv:2405.06816v14 citationsh-index: 10UAI
Originality Highly original
AI Analysis

This addresses the problem of out-of-distribution generalization for machine learning models in dynamic real-world applications, representing an incremental advance over stationary domain generalization methods.

The paper tackles domain generalization in non-stationary environments, where domains evolve over time or space, by proposing an adaptive invariant representation learning algorithm that improves performance on unseen target domains, with experimental validation on synthetic and real data.

Although recent advances in machine learning have shown its success to learn from independent and identically distributed (IID) data, it is vulnerable to out-of-distribution (OOD) data in an open world. Domain generalization (DG) deals with such an issue and it aims to learn a model from multiple source domains that can be generalized to unseen target domains. Existing studies on DG have largely focused on stationary settings with homogeneous source domains. However, in many applications, domains may evolve along a specific direction (e.g., time, space). Without accounting for such non-stationary patterns, models trained with existing methods may fail to generalize on OOD data. In this paper, we study domain generalization in non-stationary environment. We first examine the impact of environmental non-stationarity on model performance and establish the theoretical upper bounds for the model error at target domains. Then, we propose a novel algorithm based on adaptive invariant representation learning, which leverages the non-stationary pattern to train a model that attains good performance on target domains. Experiments on both synthetic and real data validate the proposed algorithm.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes