Backpropagation: how blame flows back through a layer

Set how much the loss wants each of two outputs to move: every weight gets that times its input, every input gets W transposed times it. The edges and nodes show the numbers.

This page sets no cookie and sends nothing anywhere.

Where the numbers come from

gHashTag/t27 specs/ml/layers/dense_layer.t27 (t27 master 239b4b059): every value on the page is its fn bodies, compiled to wasm (logic.js)
Course 10 of t27.ai, neural-networks (gHashTag/trinity#1695): lesson 19 opens this widget

This widget's .t27 source · All widgets