Question
The term "greedy algorithms" refers to machine-learning algorithms that:
- A. update priors as more data is seen.
- B. examine every node of a tree before making a decision.
- C. apply a theoretical model to the distribution of the data.
- D. make the locally optimal decision. Correct
Explanation
A greedy algorithm makes the choice that looks best at each step, the locally optimal decision, without revisiting earlier choices; decision-tree induction choosing the best split at each node is the classic example. Updating priors describes Bayesian methods, examining every node is exhaustive search (the opposite of greedy), and applying a theoretical distribution describes parametric modeling.
