Forward-Forward Algorithm is a new type of neural network training algorithm that's supposed to replace backpropagation. It was proposed by Geoffrey Hinton recently.

Link to the paper: https://www.cs.toronto.edu/~hinton/FFA13.pdf

Link to a Twitter thread with a less technical explanation: https://twitter.com/martin_gorner/status/1599755684941557761

Right now there is no implementation available publicly as far as I can tell, but the architecture is described in detail and it shouldn't be too difficult to implement even if the original code isn't released.

Resolves "YES" if by 2023-01-07 (UTC+0) the original code is released or the algorithm is implemented by someone else and the implemetation is released publicly (for example on GitHub). A reimplementation should achieve similar results on MNIST and CIFAR-10 to the results reported in the paper.

Dec 7, 10:39am: Will a Forward-Forward neural network implementation be publicly available before Jan 7th 2022? → Will a Forward-Forward neural network implementation be publicly available before Jan 7th 2023?