Even smartest AI models dont match human visual processing


According to Professor James Elder, co-author of a York University study published today, deep convolutional neural networks (DCNNs) don’t sense objects the same way humans do, employing configural form perception, and this could be harmful in real-world AI applications.

Deep learning algorithms fail to capture the configural structure of human shape perception is a collaborative study led by Elder, who holds the York Research Chair in Human and Computer Vision and is Co-Director of York’s Centre for AI & Society. It was published in the Cell Press journal iScience. Read More…