Autoencoder: learning a representation from unlabelled data
The network uses its own input as the target. Because it is forced through a narrow bottleneck, it has to decide what matters.
4 steps
175 XP
A free account is needed
Start the lesson →
Sources
- Hinton, G. E. & Salakhutdinov, R. R. 2006 · Reducing the Dimensionality of Data with Neural Networks · Science 313(5786)
- Baldi, P. & Hornik, K. 1989 · Neural Networks and Principal Component Analysis · Neural Networks 2(1)
- Goodfellow, I., Bengio, Y. & Courville, A. 2016 · Deep Learning, Bölüm 14 · MIT Press
ML Academy · an interactive machine learning course that runs in your browser ·
All lessons