instabilities

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.

Instabilities in a dynamic process refer to a tendency to move rapidly between very different states without settling into a stable state. It can occur in natural or artificial processes if there is positive feedback. It can also occur during the training phase of machine learning if the rate of training is too great meaning that each each training example pulls the system too far.

Used on Chap. 8: page 153