LeNet

Family of convolutional neural networks by Yann LeCun and colleagues at AT&T Bell Labs; LeNet-1 read zip-code digits around 1989, LeNet-5 read bank cheques in 1998.

LeNet is a family of convolutional neural networks (CNNs) developed by Yann LeCun and colleagues at AT&T Bell Labs. LeNet-1 around 1989 recognised handwritten zip-code digits in the Bell Labs postal experiment (article); LeNet-5 in 1998 (LeCun, Bottou, Bengio, Haffner, Gradient-Based Learning Applied to Document Recognition, Proc. IEEE 86(11), doi:10.1109/5.726791) read digits on bank cheques at commercial scale. Training relied on backpropagation through shared convolutional filters, the same recipe that later fed into AlexNet and modern deep learning.