Reading: 2.1 – 2.3 and 2.6.

1
2
3
4
5
 Dual cones and generalized inequalit
Dual cones and generalized inequalities
dual cone of a cone K:
dual cone of a cone K:
K ∗ = {y | y T x ≥ 0 for all x ∈ K}
2 Co
∗ T
K = {y | y x ≥ 0 for all x ∈ K}
examples
examples
• K = Rn+: K ∗ = Rn+ K
∗
n ∗ n K
• K = R+: K • =KR= + Sn : K ∗ = Sn
+ +
n ∗ n
• K = S+: K • =KS+ = {(x, t) | ∥x∥2 ≤ t}: K ∗ = {(x, t) | ∥x∥2 ≤ t}
• K = {(x, t) | ∥x∥2 ≤ t}: K ∗ = {(x, t) | ∥x∥2∗ ≤ t}
• K = {(x, t) | ∥x∥1 ≤PSfrag
t}:replacements
K = {(x, t) | ∥x∥∞ ≤ t}
• K = {(x, t) | ∥x∥1 ≤ t}: K ∗ = {(x, t) | ∥x∥∞ ≤ t}
first three examples are self-dual cones T
first three examples are self-dual cones If K is a halfspace, K = {x | v x ≥ 0}, the dual cone is the ray
∗

dual cones of proper cones are proper, hence 2

define generaliz
K = {tv | t ≥ 0}.
∗
dual cones of proper cones are proper, hence define
cone is the generalized
line perpendicular to v inequalities:
If K = R , the dual cone is K = {0}. If K is a line {tv | t ∈ R} through
the dual

T K = {y | v y = 0}. ∗ T
y ≽K ∗ 0 ⇐⇒
T (c) See part (a).
y x ≥ 0 for all x ≽K 0
y ≽K ∗ 0 ⇐⇒ y x ≥ 0 for all x ≽K 0 6
(d) K must be closed convex and pointed, and have nonempty interior. From
7
8
9
10
11
12
13
14
Reading: 3.1-3.5

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
 Quasiconvex functions

f : Rn → R is quasiconvex if dom f is convex and the sublevel sets

Sα = {x ∈ dom f | f (x) ≤ α}

are convex for all α

a b c

• f is quasiconcave if −f is quasiconvex
• f is quasilinear if it is quasiconvex and quasiconcave

Convex functions 3–23

35
 Examples
!
• |x| is quasiconvex on R
• ceil(x) = inf{z ∈ Z | z ≥ x} is quasilinear
• log x is quasilinear on R++
• f (x1, x2) = x1x2 is quasiconcave on R2++
• linear-fractional function
aT x + b
f (x) = T , dom f = {x | cT x + d > 0}
c x+d
is quasilinear
• distance ratio
∥x − a∥2
f (x) = , dom f = {x | ∥x − a∥2 ≤ ∥x − b∥2}
∥x − b∥2

is quasiconvex

Convex functions 3–24 36

 Log-concave and log-convex functions
a positive function f is log-concave if log f is concave:

f (θx + (1 − θ)y) ≥ f (x)θ f (y)1−θ for 0 ≤ θ ≤ 1

f is log-convex if log f is convex

• powers: xa on R++ is log-convex for a ≤ 0, log-concave for a ≥ 0

• many common probability densities are log-concave, e.g., normal:

1 1 T
Σ−1 (x−x̄)
f (x) = ! e− 2 (x−x̄)
(2π)n det Σ

• cumulative Gaussian distribution function Φ is log-concave

" x
1 2
Φ(x) = √ e−u /2 du
2π −∞

Convex functions 3–27 37

example: yield function

Y (x) = prob(x + w ∈ S)

• x ∈ Rn: nominal parameter values for product

• w ∈ Rn: random variations of parameters in manufactured product

• S: set of acceptable values

if S is convex and w has a log-concave pdf, then

• Y is log-concave

• yield regions {x | Y (x) ≥ α} are convex

Convex functions 3–30 38