Noise increases the correspondence between artificial and human vision

Journal article

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.

Authors: Thompson, JAF

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Public Library of Science
• Journal: PLoS Biology
• Volume: 19
• Issue: 12
• Pages: e3001477-e3001477
• Publication date: 2021-12-10
• DOI: 10.1371/journal.pbio.3001477
• EISSN: 1545-7885
• ISSN: 1544-9173
• Language: English
• Keywords: Noise (video); Biology; Artificial neural network; Artificial intelligence; Mirroring; Deep neural networks; Image (mathematics); Computer science; Neuroscience; Communication