Large Learning Rates Simultaneously Achieve Robustness to Spurious Correlations and Compressibility

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Barsbey, Melih, Prieto, Lucas, Zafeiriou, Stefanos, Birdal, Tolga
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912520713797632
author Barsbey, Melih
Prieto, Lucas
Zafeiriou, Stefanos
Birdal, Tolga
author_facet Barsbey, Melih
Prieto, Lucas
Zafeiriou, Stefanos
Birdal, Tolga
contents 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 facilitator for simultaneously achieving robustness to spurious correlations and network compressibility. We demonstrate that large learning rates also produce desirable representation properties such as invariant feature utilization, class separation, and activation sparsity. Our findings indicate that large learning rates compare favorably to other hyperparameters and regularization methods, in consistently satisfying these properties in tandem. In addition to demonstrating the positive effect of large learning rates across diverse spurious correlation datasets, models, and optimizers, we also present strong evidence that the previously documented success of large learning rates in standard classification tasks is related to addressing hidden/rare spurious correlations in the training dataset. Our investigation of the mechanisms underlying this phenomenon reveals the importance of confident mispredictions of bias-conflicting samples under large learning rates.
format Preprint
id arxiv_https___arxiv_org_abs_2507_17748
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Large Learning Rates Simultaneously Achieve Robustness to Spurious Correlations and Compressibility
Barsbey, Melih
Prieto, Lucas
Zafeiriou, Stefanos
Birdal, Tolga
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
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 facilitator for simultaneously achieving robustness to spurious correlations and network compressibility. We demonstrate that large learning rates also produce desirable representation properties such as invariant feature utilization, class separation, and activation sparsity. Our findings indicate that large learning rates compare favorably to other hyperparameters and regularization methods, in consistently satisfying these properties in tandem. In addition to demonstrating the positive effect of large learning rates across diverse spurious correlation datasets, models, and optimizers, we also present strong evidence that the previously documented success of large learning rates in standard classification tasks is related to addressing hidden/rare spurious correlations in the training dataset. Our investigation of the mechanisms underlying this phenomenon reveals the importance of confident mispredictions of bias-conflicting samples under large learning rates.
title Large Learning Rates Simultaneously Achieve Robustness to Spurious Correlations and Compressibility
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.17748