local search

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Local search is about finding the best point within a region of the fitness landscape. It is about fine tuning the parameters of a model. It is about fine tuning the parameters of a model. Techniques such as hill climbing or gradient descent for continuous paramters work well at this scale. Local search is in contrast to global search, which is about finding the best overall region.

Defined on page 184

Used on Chap. 9: pages 184, 185