sigmoid function

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.

A sigmoid function is a generic term refering to S-like functions that have a step-like nature, but are smooth without the discontinuity of the threshold function. This maes them more emnable to use in machine learning as, in general, continuity aids learning.

Used in Chap. 6: page 87; Chap. 7: page 107; Chap. 9: pages 136, 137; Chap. 10: page 151; Chap. 21: page 357

Also known as sigmoid

Sigmoid function smoothly limits extreme values