Complexity Matters: Dynamics of Feature Learning in the Presence of
Spurious Correlations

v1v2 (latest)

Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations

5 March 2024

ArXiv (abs)PDF HTML

Papers citing "Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations"

7 / 7 papers shown

Title
Large Learning Rates Simultaneously Achieve Robustness to Spurious Correlations and Compressibility Melih Barsbey Lucas Prieto Stefanos Zafeiriou Tolga Birdal 188 0 0 23 Jul 2025
Hypernym Bias: Unraveling Deep Classifier Training Dynamics through the Lens of Class HierarchyInternational Conference on Artificial Intelligence and Statistics (AISTATS), 2025 Roman Malashin Valeria Yachnaya Alexander Mullin 330 1 0 17 Feb 2025
Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization Simone Bombari Marco Mondelli 594 5 0 03 Feb 2025
OOD-Chameleon: Is Algorithm Selection for OOD Generalization Learnable? Liangze Jiang Damien Teney OODD OOD 467 1 0 03 Oct 2024
Accuracy on the wrong line: On the pitfalls of noisy data for out-of-distribution generalisation Amartya Sanyal Yaxi Hu Yaodong Yu Yian Ma Yixin Wang Bernhard Schölkopf OODD 168 7 0 27 Jun 2024
Learned feature representations are biased by complexity, learning order, position, and more Andrew Kyle Lampinen Stephanie C. Y. Chan Katherine Hermann AI4CE FaML SSL OOD 211 18 0 09 May 2024
How Spurious Features Are Memorized: Precise Analysis for Random and NTK FeaturesInternational Conference on Machine Learning (ICML), 2023 Simone Bombari Marco Mondelli AAML 337 9 0 20 May 2023