Improving Generalization with Domain Convex Game

by   Fangrui Lv, et al.
Beijing Institute of Technology

Domain generalization (DG) tends to alleviate the poor generalization capability of deep neural networks by learning model with multiple source domains. A classical solution to DG is domain augmentation, the common belief of which is that diversifying source domains will be conducive to the out-of-distribution generalization. However, these claims are understood intuitively, rather than mathematically. Our explorations empirically reveal that the correlation between model generalization and the diversity of domains may be not strictly positive, which limits the effectiveness of domain augmentation. This work therefore aim to guarantee and further enhance the validity of this strand. To this end, we propose a new perspective on DG that recasts it as a convex game between domains. We first encourage each diversified domain to enhance model generalization by elaborately designing a regularization term based on supermodularity. Meanwhile, a sample filter is constructed to eliminate low-quality samples, thereby avoiding the impact of potentially harmful information. Our framework presents a new avenue for the formal analysis of DG, heuristic analysis and extensive experiments demonstrate the rationality and effectiveness.


Learning to Diversify for Single Domain Generalization

Domain generalization (DG) aims to generalize a model trained on multipl...

An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization

A popular assumption for out-of-distribution generalization is that the ...

SRCD: Semantic Reasoning with Compound Domains for Single-Domain Generalized Object Detection

This paper provides a novel framework for single-domain generalized obje...

When Neural Networks Fail to Generalize? A Model Sensitivity Perspective

Domain generalization (DG) aims to train a model to perform well in unse...

ADRMX: Additive Disentanglement of Domain Features with Remix Loss

The common assumption that train and test sets follow similar distributi...

Style-Hallucinated Dual Consistency Learning: A Unified Framework for Visual Domain Generalization

Domain shift widely exists in the visual world, while modern deep neural...

Pseudo-Trilateral Adversarial Training for Domain Adaptive Traversability Prediction

Traversability prediction is a fundamental perception capability for aut...

Please sign up or login with your details

Forgot password? Click here to reset