This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Mathematics 105 — Spring 2004 — M. Christ 1 Solutions to selecta from problem set #2, with problem set #3 25 . Let f ( x, y ) = x  y  / p x 2 + y 2 for ( x, y ) 6 = (0 , 0), and let f (0 , 0) = 0. Show that f is not differentiable at 0. (Note: It is customary to write 0 as shorthand for (0 , , ··· , 0) ∈ R n , just as the additive identity element of a vector space is customarily denoted by 0.) Solution. Suppose that f were differentiable at 0. We have f (0+ t, 0) f (0 , 0) ≡ 0. Taking v = ( t, 0) = te 1 in the definition of Df (0) would give us 0 = lim  t →  f (0 + t, 0) f (0 , 0) T ( te 1 )   te 1  = lim t →  T ( e 1 )  1 =  T ( e 1 )  , so T ( e 1 ) = 0. Likewise f (0 , 0 + t ) = f (0 , 0) ≡ 0, so the same reasoning gives T ( e 2 ) = 0. Since e 1 , e 2 together form a basis for R 2 , this implies that T = 0, that is, T ( v ) = 0 for every v ∈ R 2 . Now consider v = ( t, t ). We are still assuming that Df (0) exists, and we have proved that if it does, it must equal 0. Therefore 0 = lim  t →  f ( t, t ) f (0 , 0) T ( t, t )   t  √ 2 = lim  t → 2 1 / 2  f ( t, t )   t  = lim  t → 2 1 / 2 t  t  , which is a contradiction. Therefore Df (0) cannot exist. 27 . Let f : R 2 → R satisfy  f ( x )  ≤  x  2 . Show that f is differentiable at 0. Solution. This is one of those cases where it’s easier to prove more than is asked for: We’ll show that Df (0) = 0. (As is so often the case in vector calculus, the notation in this equation is imperfect; the zero on the lefthand side of the equation is the origin in R 2 , while the zero on the righthand side refers to the linear transformation T : R 2 → R which satisfies T ( v ) = 0 for all v ∈ R 2 .) Letting T be the linear transformation 0, for any x 6 = 0 we have  f ( x ) f (0) T ( x )   x  =  f ( x )   x  ≤  x  . This certainly tends to 0 as x → 0, so we’ve shown that Df (0) = 0. 215(a) . Define a function det : ( R n ) n → R by regarding an element x = ( x 1 , ··· , x n ) ∈ ( R n ) n as a matrix whose rows are the vectors x j , and taking its determinant. Show that det is a differentiable function, and show that D det( a )( x ) = n X i =1 det ( a 1 , a 2 , ··· , a i 1 , x i , a i +1 , ··· , a n ) . (Here ( a 1 , a 2 , ··· , a i 1 , x i , a i +1 , ··· ) ∈ ( R n ) n , so we can form its determinant. Note also that the notation (which is commonly used) is slightly misleading when i = 1 or n .) Comment. In this problem we systematically identify ( R n ) n with R n 2 via the corre spondence (( x 1 1 , ··· , x n 1 ) , ··· , ( x 1 n , ··· , x n n )) ↔ ( x 1 1 , ··· , x n 1 , x 1 2 , ··· , x n 2 , x 1 3 , ··· ). And we also identify elements of ( R n ) n...
View
Full
Document
This note was uploaded on 05/07/2008 for the course MATH 105 taught by Professor Michaelchrist during the Spring '04 term at Berkeley.
 Spring '04
 MichaelChrist
 Math

Click to edit the document details