Chapter 6 Architecture and Equilibria 6.3 Synaptic convergence to centroids:AVQ Algor ithms Differential Competitive Learning (DCL) m(t+1)=m(t)+c△S(y(t)[x(t)-m(t)]6-18 mi(t+1)=mi(t)fi≠j 6-19 AS(())denotes the time change of the jth neuron's competitive signal.In practice we only use the sign of(6-20) △Sy,(t+1)=S((t+1)-S(t) 6-20 Stochastic Equilibrium and Convergence Competitive synaptic vector coverge to decsion-class centrols. May coverge to locally maxima. 2004.11.10 132004.11.10 13 Chapter 6 Architecture and Equilibria 6.3 Synaptic convergence to centroids:AVQ Algorithms Differential Competitive Learning (DCL) ( 1) ( ) 6 19 ( 1) ( ) ( ( ))[ ( ) ( )] 6 18 + =  − + = +  − − m t m t if i j m t m t c S y t x t m t i i j j t j j j S (y (t +1)) j j denotes the time change of the jth neuron’s competitive signal . In practice we only use the sign of (6-20) Sj(yj (t +1)) = Sj(yj(t +1)) − Sj(yj(t)) 6− 20 Stochastic Equilibrium and Convergence Competitive synaptic vector coverge to decsion-class centrols. May coverge to locally maxima