On the Fly Neural Style Smoothing for Risk-Averse Domain Generalization

TL;DR

This paper introduces Test-Time Neural Style Smoothing (TT-NSS), a method for risk-averse domain generalization that stylizes test images on the fly and abstains from uncertain predictions, improving reliability in unseen domains.

Contribution

It proposes a novel inference procedure and training method that enhance risk-averse predictions for domain generalization classifiers, requiring only black-box access and integrating seamlessly with existing methods.

Findings

TT-NSS improves prediction reliability on unseen domains.

The style smoothing training enhances classifier consistency.

Empirical results show improved risk-averse performance on benchmarks.

Abstract

Achieving high accuracy on data from domains unseen during training is a fundamental challenge in domain generalization (DG). While state-of-the-art DG classifiers have demonstrated impressive performance across various tasks, they have shown a bias towards domain-dependent information, such as image styles, rather than domain-invariant information, such as image content. This bias renders them unreliable for deployment in risk-sensitive scenarios such as autonomous driving where a misclassification could lead to catastrophic consequences. To enable risk-averse predictions from a DG classifier, we propose a novel inference procedure, Test-Time Neural Style Smoothing (TT-NSS), that uses a "style-smoothed" version of the DG classifier for prediction at test time. Specifically, the style-smoothed classifier classifies a test image as the most probable class predicted by the DG classifier on random re-stylizations of the test image. TT-NSS uses a neural style transfer module to stylize a test image on the fly, requires only black-box access to the DG classifier, and crucially, abstains when predictions of the DG classifier on the stylized test images lack consensus. Additionally, we propose a neural style smoothing (NSS) based training procedure that can be seamlessly integrated with existing DG methods. This procedure enhances prediction consistency, improving the performance of TT-NSS on non-abstained samples. Our empirical results demonstrate the effectiveness of TT-NSS and NSS at producing and improving risk-averse predictions on unseen domains from DG classifiers trained with SOTA training methods on various benchmark datasets and their variations.