Research · October 1986
Placed by the contemporary primary publication. The exact event date is not known; its documented interval appears below.
A method for training multilayer networks was described.
It supplied a practical training rule that later neural-network work could reuse.
The paper reports experiments; it does not resolve every difficulty of deep learning.
Nature · Published October 1986
A recurrent architecture is trained by the gradient method popularised earlier.
Both build multilayer models; the way layers are selected and trained differs.
Both describe training multilayer models by propagating error backwards; the later publication drew the wider attention.
Both concern training deep networks; this one adds layer-by-layer initialisation before fine-tuning.
It is the first rule for training hidden layers, preceding the 1986 paper from the same group.
The two volumes set out the full connectionist programme; the Nature paper of the same year is only its core.
Parallel Distributed Processing: Explorations in the Microstructure of CognitionThe network is trained by the back-propagation described a year earlier.
Parallel Networks that Learn to Pronounce English TextThe network is trained by back-propagation.
Backpropagation Applied to Handwritten Zip Code RecognitionIt shows why the 1986 method does not work at depth.
Untersuchungen zu dynamischen neuronalen NetzenThe evaluation function is a network trained by back-propagation.
Practical Issues in Temporal Difference LearningThe model is trained by back-propagation.
A Neural Probabilistic Language ModelReplacing the sigmoid removes the gradient decay inherent in the 1986 method.
The state backed the school that grew around the authors of back-propagation.
Work by the same author that gave a practical way to train deep networks.