positive feedback

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.

Positive feedback is when feeback acts in the same dircetion as changes, leading to escalating change. A familar example is the feedback effects when a micropohne picks up sound from speakers. Typically positive feedback leeds to runaway behaviours and exponantial growth, but it can be useful to create deliberate rapid change so long as there are additional mechanisms to limit the extremes. Posituve feedback effects often lead to long-tail distributions and power laws, and this can be seen in social networks when those with large numbers of connections are most likely to gain new ones.

Used in Chap. 22: page 371; Chap. 23: page 391