4.
A: Convex Set
Because all five points can be joined by a line segment and that line lies completely inside the set.
B: Non-Convex Set
Because two points of a star can be joined by the line segment which lies outside the set.
C:Non-Convex Set
As the set has an indent in which the points X and Y can be joined by the line segment which does not lie
completely inside the set.
5.
d
f ( x )=cf ( x )
dx
f ' ( x )=cf ( x )
d ' d
( f ( x ) )= ( cf ( x ) )
dx dx
d ' d
( f ( x ) )=c ( f ( x ) )
dx dx
d '
( f ( x ) )=c ( cf ( x ) )
dx
d2 ( ) 2 ( )
f x =c f x
d x2
As , c2 >0.
d2
So the > 0 , for all c . Thus, f ( x ) is convex set.
d x2
6.
Is set of rational numbers convex set?
A set c is called convex if ∀ x , y ∈C and all t ∈(0,1), the point ( 1−t ) x+ty always belongs to C.
a
A={ ∨a , b=±1 , ± 2, … }
b
3 −3 1
x= ; y= ; t=
4 4 2
1
( 1− )∗3 3 ∗1
Then 2 4
( 1−t ) x+ty= −
4 2
1 1
∗3 ∗3
2 2
− =0
4 4
A is not convex.
7.
Is set of Real numbers convex set?
A= {ab ∨a ,b ∈ R , b≠ 0}is convex
Take,b=1 , then
A1= { a|a ∈ R }=R
Thus, R=A 1 ≤ A ≤ R ↔ A=R
Clearly R is convex, hence A is convex.
8.
If following functions are convex, concave or neither?
a. f ( x )=4 x−2
f ' ' ( x ) =0
f (−x )=−4 x−2
f ' ' ( x ) =0
Function is both convex and concave.
b. f ( x )=log ( x ) , x >0
−1
f '' ( x) = 2
(ln 10)x
f ' ' ( x ) < 0; for x >0
−f ( x )=−log ( x ) , x> 0
1
−f '' ( x ) = 2
(ln 10)x
f ' ' ( x ) > 0; for x >0
So, the function is concave.
c. f ( x )=e x , x >0
f ' ' ( x ) =e x
f '' ( x) > 0
So, the function is convex.
d. f ( x )=e−x , x >0
f ' ' ( x ) =e−x , x >0
f '' ( x) > 0
So, the function is convex.
e. f ( x )=cos ( x )
f ' ' ( x ) =−cos ( x )
π 3π
f ' ' ( x ) > 0; for only +2 kπ < x < +2 kπ
2 2
So, function is neither convex nor concave.
f. f ( x )=tan ( x )
2 sin x
f ' ' ( x ) =2 sec 2 x tan x=
cos 3 x
π 3π
f ' ' ( x ) > 0; for only 0< x< ∪ π < x<
2 2
So, function is neither convex nor concave.
π
g. f ( x )=tan ( x ) ; for− < x<0
2
'' 2
f ( x ) =2 sec x tan x
π
for− < x< 0
2
''
f ( x ) < 0.
−f ( x )=−tan x
π
for− < x< 0
2
''
−f ( x ) > 0.
π
So, the function is concave for− < x< 0
2
h. f ( x )=exp ( x2 ) , for x <0
2
f ' ' ( x ) =2 e x ( 2 x 2+ 1)
f ' ' ( x ) > 0; for x <0
So, the function is convex.
π
i. f ( x )=exp ( cos ( x ) ) ; 0< x<
2
f ( x ) =−e cos x+ e sin 2 x
'' cos x cos x
f ' ' ( x ) =e cos x (sin2 x−cos x)
f ' ' ( x ) =e cos x (1−cos 2 x−cos x)
f ' ' ( x ) =e cos x ¿
π
f ' ' ( x ) > 0; for 0< x <
2
So, the function is convex.
π π
j. f ( x )=cos x ; for− < x<
2 2
''
f ( x ) =−cos x
−f ( x )=−cos x ;
−f '' ( x ) =cos x
π π
for− < x<
2 2
''
−f ( x ) > 0
So, the function is concave.
k. f ( x )=3 x 3
f ' ' ( x ) =18 x
f ' ' ( x ) > 0; for x >0
−f ( x )=−3 x 3
f ' ' (−x )=−18 x
f ' ' (−x )> 0; for x <0
So the function is neither convex nor concave
l. f ( x )=log ( x 2+ 1)
2∗(1−x 2)
f '' ( x) = 2
ln 10∗( x 2 +1 )
f ' ' ( x ) > 0; for x >1
f ' ' ( x ) < 0; for x <1
So, the function is neither convex nor concave
9.
f ( x )∧g ( x ) are convex function. So,
f '' ( x) > 0
g' ' ( x ) >0
If ( ( x )+ g ( x ) is convex.)
d
( f ( x ) + g ( x )) =f ' ( x )+ g ' ( x )
dx
d
( f ¿¿ ' ( x )+ g ' ( x ))=f ' ' ( x ) + g' ' ( x ) ¿
dx
As,
f '' ( x) > 0
g' ' ( x ) >0
So, ( f ¿ ¿ ' ' ( x )+ g ' ' ( x ) )>0 ¿
Thus f ( x ) + g ( x ) is convex.
10.
Any function h : x → y is convex if
h ( t x 1 + ( 1−t ) x2 ) ≤ th ( x 1 ) + ( 1−t ) h ( x 2 )
∀ x1 , x 2 ∈ X ,t ∈[0,1]
Now, if f and g are convex. So, these two satisfies these.
Now, consider,
( f ∘ g)(t x 1+ (1−t ) x 2)
t ∈[0,1]
g ( t x 1 + ( 1−t ) x 2 ) ≤tg ( x 1 ) +(1−t) g (x2 )
f is monotonic. So, for x ≤ y → f ( x ) ≤ f ( y )
f ( g (t x 1+ ( 1−t ) x 2) ) ≤ f ( tg ( x 1 ) + ( 1−t ) g ( x 2 ) )
f ( g (t x 1+ ( 1−t ) x 2) ) ≤ tf ( g ( x 1 ) ) +(1−t) f ( g( x 2))
( f ∘ g ) ( t x 1+ (1−t ) x 2 ) ≤ t ( f ∘ g ) ( x 1) +(1−t ) ( f ∘ g ) ( x 2)
So, ( f ∘ g ) is convex.
11.
If f ( x )is convex. Then
f '' ( x) > 0
g ( x )=−f ( x ) is concave.
Let, α ∈[0,1]
g ( ( 1−α ) x+ αy )=−f ( ( 1−α ) x+ αy )
Since f is convex,
f ( ( 1−α ) x +αy) ≤ (1−α ) f (x)+ αf ( y )
−f ( (1−α ) x +αy ) ≥−( 1−α ) f ( x)+ αf ( y )
−f ( (1−α ) x +αy ) ≥ ( 1−α ) (−f ( x ) )+ α (−f ( y ) )
−f ( (1−α ) x +αy )=( 1−α ) (g ( x ))+α ( g ( y ))
So,
g ( ( 1−α ) x+ αy ) ≥ ( 1−α ) (g ( x ))+ α ( g ( y ) )
So, g is concave.
12.
a. Minimize x 3−2 x 2+2 subject to x >2
f ( x )=x 3−2 x 2+ 2
g ( x )=x−2>0
f is convex and g is concave.
By definition.
f ( λx+ ( 1−λ ) y ) ≤ λ f ( x ) +(1−λ)f ( y)
2 2
λ 3 x 3+ ( 1−λ ) y 3 +3 λ2 x 2 ( 1−λ ) y +3 λ x ( 1−λ ) y 2−2 λ2 x2 −2 ( 1−λ ) y 2−2 λ (1−λ ) xy +2≤ λ ( x 3−2 x 2+ 2 ) +(1−λ)( y 3−
So, f is convex
g is both convex and concave. So, it is a convex optimization.
b.
1
Minimize x 2+ 2 subject to cos x ≤
2
f ( x )=x 2 +2
1
g ( x )=cos x− ≤ 0
2
2
f ( λ x+ ( 1−λ ) y )=( λ x + ( 1− λ ) y ) + 2
f ( λx+ ( 1−λ ) y )=λ 2 x 2 + ( 1−x )2 y 2 +2 λ ( 1−λ ) xy +2
λ f ( x )= λ ( x 2+2 )
( 1−λ ) f ( y )=( 1−λ)( y 2 +2)
f ( λx+ ( 1−λ ) y ) ≤ λ f ( x ) + ( 1− λ ) f ( y )
For y,
1
g ( λx + ( 1−λ ) y ) =cos ( λ x +(1−λ) y )−
2
1
(
λ g ( x )=λ cos x−
2 )
1
( 1−λ ) g ( y )=(1− λ)( cos y− )
2
So,
g ( λx + ( 1−λ ) y ) ≤ λg ( x ) + ( 1− λ ) g ( y )
So, we know that
f ( x ) is convex and g ( x ) is also convex. Thus,, the problem is convex optimization.