22. The Final Design Experiment Generator New problem (initial game board) Hypothesis Training examples Solution trace (game history) Figure 1.1 Final design of the checkers learning program Performance System Critic Generalizer { <b 1 ,V train (b 1 )> , <b 2 ,V train (b 2 )> …}
23. Choices in designing the checkers learning problem Linear programming Polynomial Linear function of six features Table of correct moves Games against self Games against experts Determine Type of Training Experience Determine Target Function Determine Representation of Learned Function Completed Design Determine Learning Algorithm … … … … Board move Board value Artificial neural network Gradient descent