Lecture 3. 4 The derivative of a map between vector spaces Let f : V → W be a smooth map between real vector spaces. Deﬁnition 4.1. Given x ∈ V we say that f is differentiable at x if there is a linear map L x : V → W so that for all v ∈ V we have: � f (x ) − f (x � ) − L x (x − x � )� = o(�x − x � �). Here we using the Landau symbol o to mean a function o : R+ → R continu ous at zero and o(0) = 0. Really this is an improper deﬁnition. We really need V and W to be normed vector spaces and it is natural to require that L is a continuous linear map. One can try to develop differential calculus on manifolds modelled on general topolog ical vector spaces. A sufﬁciently general context to work in is that of manifolds modelled on Banach spaces, that is complete normed linear spaces. Essentially of the basic results in differential topology work in this context with the same proofs (as long as proof don’t use coordinates) Notice that map L in the above deﬁnition is unique. If L � is another such map then o(�x − x � �) ≥ � f (x ) − f (x � ) − L(x − x � ) − ( f (x) − f (x � ) − L � (x − x � ))� = �(L − L � )(x − x � )� So (L − L � )(x − x � ) = 0. The map L is called the differential of f at x and is denoted dx f or Dx f. We say f is differentiable if f is differentiable at each x ∈ U and is continuously differentiable if d f : U → hom(V , W ). is continuous. The second derivative is the derivative of the ﬁrst derivative and thus is a map d 2 f : U → hom(V , hom(V , W )). 7 In the ﬁnite dimensional case hom(V, hom(V, W )) with a subspace of hom(V ⊗ V, W ) . In the inﬁnite dimensional case we need to be more careful but we can identify hom(V, hom(V, W )) with bilinear maps from V → W. You can read all about this in gory detail in [?] Deﬁnition 4.2. A smooth map f : X → Y is called an immersion its differential is everywhere injective. It is called a submersion if it differential is everywhere surjective. There obvious examples of such maps. Suppose m < n are positive integers i : Rm → R n given by i(x 1 , . . . , xm ) = (x 1 , . . . , xm , 0, . . . , 0) is an immersion while s : Rn → R m given by s(x 1 , . . . , xm , xm+1 , . . . , xn ) = (x 1 , . . . , xm ) is a submersion. We will see in the next section that locally these simple examples are completely general. 8

18.966 Geometry of Manifolds: Lecture 3

Related Documents

Recommended Documents

Description

Related Documents

Free up your schedule!

Take 5 seconds to unlock