no code implementations • 21 Sep 2023 • Jonas Brenig, Radu Timofte
This may be caused by a combination of factors, including having a loss function for each layer and the way the supervised training is realized in the forward-forward paradigm.
Representation Learning Self-Supervised Learning