Quasi Newton PDF
Quasi Newton PDF
Quasi Newton PDF
2. Quasi-Newton methods
• quasi-Newton methods
• BFGS update
2-1
Newton method for unconstrained minimization
minimize f (x)
Newton method
x+ = x − tH −1∇f (x)
∆x = −H −1∇f (x)
T
1/2
kzkH = z Hz
BFGS update
yy T Hk−1ssT Hk−1
Hk = Hk−1 + T −
y s sT Hk−1s
where
s = x(k) − x(k−1), y = ∇f (x(k)) − ∇f (x(k−1))
inverse update
T T
ssT
sy ys
Hk−1 = I− T −1
Hk−1 I − T + T
y s y s y s
T
T T
(sT v)2
s v s v
v T Hk−1v = v− T y −1
Hk−1 v− T y + T
s y s y y s
′ (k)
f (x ) f ′(x(k)) − f ′(x(k−1))
x(k+1) (k)
=x − , Hk =
Hk x(k) − x(k−1)
′
fquad (z)
f ′(z)
global result
local convergence
0 0
10 10
f (x(k)) − f ⋆
f (x(k)) − f ⋆
-2 -2
10 10
-4 -4
10 10
-6 -6
10 10
-8 -8
10 10
-10 -10
10 10
-12 -12
10 10
0 1 2 3 4 5 6 7 8 9 0 20 40 60 80 100 120 140
k k
T
(αỹ − s̃) s̃
Lk = Lk−1 I + ,
s̃T s̃
where 1/2
T
s̃ s̃
ỹ = L−1
k−1 y, s̃ = Lk−1s, α=
yT s
−1 −1
minimize tr(Hk−1 X) − log det(Hk−1 X) − n
subject to Xs = y
1 T
X −1
= −1
Hk−1 − (sν + νsT ), Xs = y, X≻0
2
with ! !
T −1
1 −1 y Hk−1 y
ν= 2Hk−1 y − 1+ s
sT y yT s
−1
for j = k, k − 1, . . . , k − m + 1, assuming, for example, Hk−m =I
• cost per iteration is O(nm); storage is O(nm)