Skip to main navigation Skip to search Skip to main content

Learning Robust Classifiers with Self-Guided Spurious Correlation Mitigation

  • University of Virginia

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

Deep neural classifiers tend to rely on spurious correlations between spurious attributes of inputs and targets to make predictions, which could jeopardize their generalization capability. Training classifiers robust to spurious correlations typically relies on annotations of spurious correlations in data, which are often expensive to get. In this paper, we tackle an annotation-free setting and propose a self-guided spurious correlation mitigation framework. Our framework automatically constructs fine-grained training labels tailored for a classifier obtained with empirical risk minimization to improve its robustness against spurious correlations. The fine-grained training labels are formulated with different prediction behaviors of the classifier identified in a novel spuriousness embedding space. We construct the space with automatically detected conceptual attributes and a novel spuriousness metric which measures how likely a class-attribute correlation is exploited for predictions. We demonstrate that training the classifier to distinguish different prediction behaviors reduces its reliance on spurious correlations without knowing them a priori and outperforms prior methods on five real-world datasets.

Original languageEnglish
Title of host publicationProceedings of the 33rd International Joint Conference on Artificial Intelligence, IJCAI 2024
EditorsKate Larson
PublisherInternational Joint Conferences on Artificial Intelligence
Pages5599-5607
Number of pages9
ISBN (Electronic)9781956792041
StatePublished - 2024
Event33rd International Joint Conference on Artificial Intelligence, IJCAI 2024 - Jeju, Korea, Republic of
Duration: Aug 3 2024Aug 9 2024

Publication series

NameIJCAI International Joint Conference on Artificial Intelligence
ISSN (Print)1045-0823

Conference

Conference33rd International Joint Conference on Artificial Intelligence, IJCAI 2024
Country/TerritoryKorea, Republic of
CityJeju
Period08/3/2408/9/24

Fingerprint

Dive into the research topics of 'Learning Robust Classifiers with Self-Guided Spurious Correlation Mitigation'. Together they form a unique fingerprint.

Cite this