AI’s dangerous shortcuts: Deep convolutional neural networks insensitive to configurational object properties

AI Could be Using Dangerous Shortcuts to Solve Complex Recognition Tasks

Researchers found that deep convolutional networks are insensitive to object configural properties.

Deep convolutional networks (DCNNs), which are based on configural shape perception, do not perceive things the same as humans (which could be detrimental in real-world AI). According to Professor James Elder who is the co-author of a York University paper published recently in iScience, DCNNs do not perceive things like humans (through configural shape perception). This could be harmful for real-world AI applications.

The study conducted by Elder who is the York Research Chair for Human and Computer Vision and Co-Director at York’s Centre for AI & Society and Nicholas Baker an assistant psychology professor in Chicago and a previous VISTA postdoctoral Fellow at York finds that deep learning fails to capture the configural aspect of human shape perception.

Source:

AI Use Potentially Dangerous “Shortcuts” To Solve Complex Recognition Tasks