factual

Randomness escapes local optima in search

Optimization algorithms searching for approximate solutions to complex problems use randomness, especially early on, because without it they get stuck at a local optimum—a blind alley that's good but not great—and an injected random jump lets them escape, which parallels how human creativity benefits from disruption.

factualpending

Speaker

Tim Harford

Evidence Quote

“if you throw randomness in and the computer suddenly makes a big move and tries something totally different, then it gets unstuck.”

Source

Tim Harford on the Virtues of Disorder and Messy— EconTalk
Created: 6/13/2026, 12:25:48 AM

Your notes

Loading notes...