Noise increases the correspondence between artificial and human vision.

Thompson JAF.

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.

DOI

10.1371/journal.pbio.3001477

Type

Journal article

Publication Date

2021-12-01T00:00:00+00:00

Volume

19

Keywords

Humans, Neural Networks, Computer

Permalink More information Close