PT - JOURNAL ARTICLE AU - Li, Zhe AU - Caro, Josue Ortega AU - Rusak, Evgenia AU - Brendel, Wieland AU - Bethge, Matthias AU - Anselmi, Fabio AU - Patel, Ankit B. AU - Tolias, Andreas S. AU - Pitkow, Xaq TI - Robust deep learning object recognition models rely on low frequency information in natural images AID - 10.1101/2022.01.31.478509 DP - 2022 Jan 01 TA - bioRxiv PG - 2022.01.31.478509 4099 - http://biorxiv.org/content/early/2022/02/02/2022.01.31.478509.short 4100 - http://biorxiv.org/content/early/2022/02/02/2022.01.31.478509.full AB - Machine learning models have difficulty generalizing to data outside of the distribution they were trained on. In particular, vision models are usually vulnerable to adversarial attacks or common corruptions, to which the human visual system is robust. Recent studies have found that regularizing machine learning models to favor brain-like representations can improve model robustness, but it is unclear why. We hypothesize that the increased model robustness is partly due to the low spatial frequency preference inherited from the neural representation. We tested this simple hypothesis with several frequency-oriented analyses, including the design and use of hybrid images to probe model frequency sensitivity directly. We also examined many other publicly available robust models that were trained on adversarial images or with data augmentation, and found that all these robust models showed a greater preference to low spatial frequency information. We show that preprocessing by blurring can serve as a defense mechanism against both adversarial attacks and common corruptions, further confirming our hypothesis and demonstrating the utility of low spatial frequency information in robust object recognition.Competing Interest StatementThe authors have declared no competing interest.