Deep Residual Learning for Image Recognition
At a glance
ResNet, an image-recognition network, uses shortcuts between layers to make much deeper networks easier to train.
In one paragraph
Introduced ResNet, using shortcut connections so layers learn refinements to information passed from earlier layers. This made much deeper networks easier to train. The paper demonstrated 152-layer networks for image recognition, with ResNet-based systems winning major vision competitions in 2015.
