Convolutional Neural Networks

Vision transformers have bested convolutional neural networks (CNNs) in a number of key vision tasks. Have CNNs hit their limit? New research suggests otherwise.

The ability to generate realistic images without waiting would unlock applications from engineering to entertainment and beyond. New work takes a step in that direction.

TinyML shows promise for bringing deep learning to applications where electrical power is scarce, processing in the cloud is impractical, and/or data privacy is paramount.

Work on vision transformers exploded in 2022. Researchers published well over 17,000 ViT papers during the year. A major theme: combining self-attention and convolution.