A function maps inputs to outputs. In machine learning, a model is a function with unknown parameters whose values are estimated from data, allowing it to approximate patterns rather than solve only one fixed task. section 5.1
From an ellipse to a general model
Kepler tested candidate models against observations and settled on an ellipse. A neural network could instead approximate a wider family of functions from input-output pairs, adapting to the task through learned parameters. section 5.1
Figure 5.1 Johannes Kepler considers multiple candidate models that might fit the data at hand, settling on an ellipse.
Compare candidate models fitted to observations, including the ellipse selected as a useful model. section 5.1
