Noise increases the correspondence between artificial and human vision

Thompson, Jessica A. F. (2021) Noise increases the correspondence between artificial and human vision. PLOS Biology, 19 (12). e3001477. ISSN 1545-7885

Text
journal.pbio.3001477.pdf - Published Version
Download (502kB)

Official URL: https://doi.org/10.1371/journal.pbio.3001477

Abstract

The best performing computer vision systems are based on deep neural networks (DNNs). A study in this issue of PLOS Biology shows that DNNs trained on noisy stimuli are better than standard DNNs at mirroring both human behavioral and neural visual responses.

Item Type:	Article
Subjects:	Euro Archives > Biological Science
Depositing User:	Managing Editor
Date Deposited:	15 Mar 2023 08:23
Last Modified:	28 Mar 2024 03:33
URI:	http://publish7promo.com/id/eprint/1281

Actions (login required)

: View Item