Matakuliah Tahun Versi : H0434/Jaringan Syaraf Tiruan : 2005 :1 Pertemuan 17 HOPFIELD NETWORK 1 Learning Outcomes Pada akhir pertemuan ini, diharapkan mahasiswa akan mampu : • Menjelaskan konsep dari Jaringan Hopfield 2 Outline Materi • Hopfield Model. • Lyapnov Function. 3 Hopfield Model 4 Equations of Operation dn i(t) C ------------- = dt ni ai C Ii T i j 1 = --------R i j S j =1 n (t ) T i j a j(t ) – ----i------ + Ii Ri - input voltage to the ith amplifier - output voltage of the ith amplifier - amplifier input capacitance - fixed input current to the ith amplifier 1 1 ----- = --- + Ri S 1 --R------- j=1 i j ni = f –1 a i a i = f ni 5 Network Format dn i(t) Ri C ------------- = dt S Ri T i ja j(t) – ni(t ) + RiI i j =1 Define: = RiC w i j = RiT i j d ni(t ) ------------- = – n i(t) + dt b i = R iI i S wi j aj(t ) + bi j=1 Vector Form: d n(t) ------------ = – n(t) + Wa(t ) + b dt a(t ) = f n(t) 6 Hopfield Network 7 Lyapunov Function ai T 1 T V a = – --- a Wa + f –1 u du – b a 2 i = 1 0 S 8 Individual Derivatives First Term: T da d 1 T 1 T T da T da – -a Wa = – a Wa ------ = – Wa ------ = – a W ----- dt 2 2 dt dt dt Second Term: ai d f d t 0 ai –1 d u du = f d a i 0 da i da i da i –1 –1 = f ai = ni u du d t d t dt ai T da d ----- f –1 u du = n -----dt dt i = 1 0 S Third Term: d T T T da T da – b a = – b a ------ = – b -----dt dt dt 9 Complete Lyapunov Derivative d T da T da T da T T T da V a = – a W ------ + n ------ – b ------ = – a W + n – b -----dt dt dt dt dt From the system equations we know: d n(t ) – a W + n – b = – -----------dt T T T T So the derivative can be written: d n(t ) d V a = – -----------dt dt S = – i=1 If S T da ------ = – dt i=1 dn da i i = – d t d t S i=1 dn da i i d t d t 2 da – 1 d f a i i d t d a i –1 d f ai > 0 d ai then d V a 0 dt 10 Invariant Sets Z = a : dV a dt = 0 a in the closure ofG S 2 da d d – 1 i V a = – f a i dai dt dt i =1 This will be zero only if the neuron outputs are not changing: da ------ = 0 dt Therefore, the system energy is not changing only at the equilibrium points of the circuit. Thus, all points in Z are potential attractors: L = Z 11 Example 2 n = ----- tan --- a 2 2 –1 n a = f n = ---tan --------- 2 R1 2 = R2 1 = 1 W = 0 1 1 0 T 1 2 = T 2 1 = 1 = RiC = 1 = 1.4 I 1 = I2 = 0 b = 0 0 12 Example Lyapunov Function ai T 1 T –1 -V a = – a Wa + f u du – b a 2 i = 1 0 S a1 1 T 1 – ---a Wa = – --- a 1 a 2 0 1 = –a1a2 2 2 1 0 a2 ai f 0 ai ai 0 0 2 2 2 u du = ------ tan --- u d u = ------ – log cos --- u -- 2 2 –1 4 = – --------2- log cos --- a i 2 4 V a = – a 1 a 2 – ------------2- log cos -- a 1 + log cos --- a2 2 2 1.4 13 Example Network Equations dn ------- = – n + Wf n = – n + Wa dt dn1 dt = a2 – n1 dn2 dt = a 1 – n2 2 –1 1.4 a 1 = ---tan ----------- n1 2 2 –1 1.4 a 2 = ---tan ----------- n2 2 14 Lyapunov Function and Trajectory 1 V(a) 2 0.5 1 a2 0 0 -0.5 1 0.5 1 0.5 0 -1 -1 -0.5 0 0.5 1 a2 0 -0.5 -0.5 -1 -1 a1 a1 15 Time Response 1 2 1.5 0.5 a2 1 0 V(a) 0.5 -0.5 a1 -1 0 0 2 4 6 t 8 10 0 2 4 6 8 10 t 16 Convergence to a Saddle Point 1 0.5 a2 0 -0.5 -1 -1 -0.5 0 a1 0.5 1 17 Hopfield Attractors The potential attractors of the Hopfield network satisfy: da ------ = 0 dt How are these points related to the minima of V(a)? The minima must satisfy: V = V V ... V a1 a2 aS T = 0 Where the Lyapunov function is given by: ai T 1 V a = – --- a Wa + f –1 u du – b a 2 i = 1 0 S T 18 Hopfield Attractors Using previous results, we can show that: d n(t ) V a = – Wa + n – b = – -----------dt The ith element of the gradient is therefore: dn i da i –1 –1 d d V a = – = – ( f a i ) = – f ai dt dt d ai dt ai Since the transfer function and its inverse are monotonic d –1 increasing: f a > 0 dai All points for which d a(t) ------------ = 0 dt i will also satisfy V a = 0 Therefore all attractors will be stationary points of V(a). 19 Effect of Gain 2 – 1 n a = f n = --- tan --------- 2 1 = 14 = 1.4 0.5 = 0.14 a 0 -0.5 -1 -5 -2.5 0 n 2.5 5 20 Lyapunov Function ai T 1 T V a = – --- a Wa + f –1 u du – b a 2 i = 1 0 S ai f 0 –1 2 u du = ----- f –1 2 u u = ------ tan ------ 2 a i a i 2 4 --- log cos -------- = – --------2- log cos -------- 2 2 1.5 = 0.14 1 a 4 – --------2- log cos -------- 2 = 1.4 0.5 0 = 14 -1 -0.5 0 a 0.5 1 21 High Gain Lyapunov Function As the Lyapunov function reduces to: 1 T T V a = – --- a Wa – b a 2 The high gain Lyapunov function is quadratic: T T 1 T 1 T V a = – -- a Wa – b a = -- a Aa + d a + c 2 2 where 2 V a = A = –W d = –b c = 0 22 Example 2 V a = – W = 0 –1 –1 0 1 = –1 2V a – I = z1 = 1 – –1 –1 – 2 = – 1 = + 1 – 1 2 = 1 1 z2 = 1 –1 1 V(a) 1 0.5 0.5 a2 0 0 -0.5 -0.5 -1 1 0.5 1 0.5 0 -1 -1 -0.5 0 a1 0.5 1 a2 0 -0.5 -0.5 -1 -1 a1 23
© Copyright 2026 Paperzz