paper-with-me

홈 › Papers

An Investigation of Why Overparameterization Exacerbates Spurious Correlations

2020-05-09 · Shiori Sagawa, aditi raghunathan, Pang Wei Koh, Percy Liang

We study why overparameterization -- increasing model size well beyond the point of zero training error -- can hurt test error on minority groups despite improving average test error when there are spurious correlations in the data. Through simulations and experiments on two image datasets, we identify two key properties of the training data that drive this behavior: the proportions of majority versus minority groups, and the signal-to-noise ratio of the spurious correlations. We then analyze a linear setting and theoretically show how the inductive bias of models towards "memorizing" fewer examples can cause overparameterization to hurt. Our analysis leads to a counterintuitive approach of subsampling the majority group, which empirically achieves low minority error in the overparameterized regime, even though the standard approach of upweighting the minority fails. Overall, our results suggest a tension between using overparameterized models versus using all the training data for achieving low worst-group error.

📄 PDF Abstract BibTeX arXiv:2005.04345

Code (3)

ssagawa/overparam_spur_corr 공식 구현 pytorch
https://worksheets.codalab.org/worksheets/0x1db77e603a8d48c8abebd67fce39cf8b 공식 구현
qiaoruiyt/noiserobustdg pytorch

Tasks

Inductive Bias

Similar Papers 제목 키워드 기반

Overparameterization hurts worst-group accuracy with spurious correlations

2020-01-01 · ICML 2020 1 · Shiori Sagawa, aditi raghunathan, Pang Wei Koh, Percy Liang

Increasing model capacity well beyond the point of zero training error has been observed to improve average test accuracy. However, such overparameterized models have been recently shown to obtain low worst-group accurac…

Understanding Domain Generalization: A Noise Robustness Perspective

2024-01-26 · Rui Qiao, Bryan Kian Hsiang Low

Despite the rapid development of machine learning algorithms for domain generalization (DG), there is no clear empirical evidence that the existing DG algorithms outperform the classic empirical risk minimization (ERM) a…

Domain Generalization

Defending Deep Neural Networks against Backdoor Attacks via Module Switching

2025-04-08 · Weijun Li, Ansh Arora, Xuanli He, Mark Dras 외

The exponential increase in the parameters of Deep Neural Networks (DNNs) has significantly raised the cost of independent training, particularly for resource-constrained entities. As a result, there is a growing relianc…

Evolutionary AlgorithmsSST-2

Which Spurious Correlations Impact Reasoning in NLI Models? A Visual Interactive Diagnosis through Data-Constrained Counterfactuals

2023-06-21 · Robin Chan, Afra Amini, Mennatallah El-Assady

We present a human-in-the-loop dashboard tailored to diagnosing potential spurious features that NLI models rely on for predictions. The dashboard enables users to generate diverse and challenging examples by drawing ins…

Logical Fallacies

Large Learning Rates Simultaneously Achieve Robustness to Spurious Correlations and Compressibility

2025-07-23 · Melih Barsbey, Lucas Prieto, Stefanos Zafeiriou, Tolga Birdal arxiv

Robustness and resource-efficiency are two highly desirable properties for modern machine learning models. However, achieving them jointly remains a challenge. In this paper, we identify high learning rates as a facilita…