Kiru Lab  /  The Mathematics of Learning

Uncertainty and Loss

Probability as an honest account of ignorance, and loss functions as its consequence.

In this module you will

  • Distinguish a probability from a confidence-shaped number
  • Derive squared error and cross-entropy from likelihood
  • Explain why a model can be accurate and badly calibrated
Distributions and Expectation
A distribution is a statement about what you do not know. Expectation is its center of mass.
Concept · 25m
Where Loss Functions Come From
Squared error and cross-entropy are not arbitrary. Both fall out of maximizing likelihood.
Concept · 30m
← The Mathematics of Learning