OptimizationNew
Gradient Descent
Downhill blindfolded: feel the slope under your feet and take a small step down.
Needs:Loss Functions
Explore
Downhill blindfolded: feel the slope under your feet and take a small step down.
Summarize the relationship between two quantities with a straight line, then use it to predict.
One number that says how wrong a model is — the thing training tries to make small.
Tame an overfitting model by penalizing large weights.
How big each step is. Too small never arrives; too big jumps over the valley.
A model that memorizes its training data fails on new data.