deep-learning
A loss function -- one number for how wrong
taught
Taught in
-
Why training needs a single number
Deep Learning / Neural Networks
Requires
- error-and-squared-error required
- sums-and-averages required
Required by
- Matching a loss function to a task required
- The loss landscape as a search space required
- Train, validation, and test splits 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.