Representations of regular and irregular shapes by deep Convolutional Neural Networks, monkey inferotemporal neurons and human judgments
In conclusion, the representations of abstract shape similarity a re highly comparable between macaque IT neurons and deep convolutional layers of CNNs that were trained to classify natural images, while human shape similarity judgments correlate better with the deepest layers.
Source: PLoS Computational Biology - Category: Biology Authors: Ioannis Kalfas Source Type: research