deep-learning
Forward propagation
taught
Taught in
-
The forward pass, one layer at a time
Deep Learning / Neural Networks
-
A network is a function built from functions
Deep Learning / Neural Networks
-
A full forward pass, by hand
Deep Learning / Neural Networks
Requires
- layers-and-parameters required
-
matrix-multiplication
required
A layer's forward pass is a matrix-vector product plus a bias.
Required by
- Autoencoders required
- Computational graphs required
- Linear regression as a one-layer network required
- Loss as the training signal required
- Sequential data and the autoregressive idea required
- Training versus inference required
- Why a network needs a fast way to find every weight's gradient required
Local knowledge map
Local knowledge map
Click to expand
Solid arrows are required prerequisites; dashed are recommended or helpful. Lower rows are more foundational. Click any box to jump to that concept.