MEPX
Chapter 1 of 10All chapters
  1. 1The artificial neuron
  2. 2Layers and depth
  3. 3How it learns
  4. 4Activation functions
  5. 5Architectures
  6. 6Keeping it general
  7. 7Training in practice
  8. 8Transfer learning
  9. 9What they cannot do
  10. 10Learning it yourself

Chapter 1 of 10

The artificial neuron

The unit everything is built from.

What it computes

Multiply each input by a weight, add them up with a bias, then pass the result through a non-linear function. That is the entire operation, repeated billions of times.

  • The weights are what training adjusts; the structure stays fixed.
  • The biological analogy is loose and mostly historical.

Why non-linearity

Without the non-linear step, any stack of layers collapses mathematically into a single linear function. That one detail is what makes depth meaningful.