Exercise 6.2.1
Feature vectors: motivation vs tabular, goal of feature engineering
Reinforcement Learning › Exercises › Lecture 6 · Function Approximation › 6.2 Linear function approximation
Discuss the primary motivations for representing states with feature vectors, \(\mathbf{x}(s)\), in value function approximation as opposed to using a tabular representation. What is the goal of engineering a good set of features?