Lecture 12

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: to control the contribu0on of the previous velocity by adjus0ng iner0a weight ω in the range of 0 and 1. € v t −1 = 0 v i ( t ) = c1 × r1 × ( pbest ( t − 1) − x i ( t − 1)) + c 2 × r2 × ( gbest ( t − 1) − x i ( t − 1)) v i ( t ) = ω × v i (t − 1) + c1 × r1 × ( pbest ( t − 1) − x i (t − 1)) + c 2 × r2 × (gbest ( t − 1) − x i (t − 1)) •  Large ω: facilitate explora0on, with increased diversity. •  small ω: promote local search. Iner0a Weight and Velocity • Since only a percentage of the velocity is retained from the previous itera0on, the iner0a weight may be thought of as a simplified form of fric0on, which reduces veloci0es over 0me – thus helping par0cles seUle onto a good solu0on without excessive wandering. •  In analogous to the social and cogni0ve aspects of the algorithm, one may think of the iner0a weight as controlling the rate at which par0cles for...
View Full Document

This document was uploaded on 02/28/2014 for the course CS 6776 at Memorial University.

Ask a homework question - tutors are online