Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

The question is motivated from the definition of $C^r(\Omega)$ I learned from S.S.Chern's Lectures on Differential Geometry:

Suppose $f$ is a real-valued function defined on an open set $\Omega\subset{\bf R}^m$. If all the $k$-th order partial derivatives of $f$ exist and are continuous for $k\leq r$, then we say $f\in C^r(\Omega)$. Here $r$ is some positive integer.

While in Folland's Introduction to Partial Differential Equations, $C^r(\Omega)$ denotes the space of functions possessing continuous derivatives up to order $r$ on $\Omega$, where $\Omega$ is an open subset of ${\bf R}^m$ and $k$ is a positive integer.

It's trivial to show that these two definitions are equivalent when $m=1$. So here is my question:

Are these two definitions equivalent in the higher dimensions? How to prove it?


Edit: The question was partially answered by Didier here. At least for me, I do not think this is a trivial question. For answering the question above, the key point, I think, is to answer the following question:

In the higher dimension case, say, $f:{\bf R}^n\to {\bf R}$($n\geq 2$), what's the relationship between the higher order partial derivative of $f$ and the high order derivative $f^{(k)}$?

When $1\leq k\leq 2$, $f^{(k)}$ is the gradient and the Hessian matrix respectively. The answer to the above questions in these two cases is clear. I have no idea for $k>2$. I don't know much about tensor, I'm not sure if the question is related to the topic of multilinear algebra.

share|improve this question
1  
I think something went wrong here. Look at the exponents of your ${\mathbb R}$'s. –  Christian Blatter Jun 10 '11 at 14:31
    
@Christian: Notation corrected. Thanks. –  Jack Jun 10 '11 at 16:41

2 Answers 2

up vote 2 down vote accepted

The $r$-th derivative $d^r f(x)$ of a function $f:\ {\mathbb R}^n\to{\mathbb R}$ at a point $x\in {\mathbb R}^n$ is a homogeneous polynomial of degree $r$ in the $n$ variables $X_1$, $\ldots$, $X_n$, where $X$ is a tangent vector "attached to the point $x$". The coefficients of this polynomial are the $r$-th partial derivatives of $f$ at the point $x$, and things are set up such that $x \to d^r f(x)$ is continuous as a "polynomial-valued function" iff all these partial derivatives are continuous. So the two definitions alluded to are actually the same.

In the case $r=1$ this means that $f\in C^1$ iff $\nabla f(x)$ is a continuous vector-valued function of $x$ iff the $n$ partial derivatives ${\partial f\over \partial x_k}$ are continuous.

If $f$ is vector-valued then the above applies to every component $f_i$ of $f$; but of course which enough sophistication we can forget about coordinates altogether.

share|improve this answer

Yes they are, see here.

share|improve this answer
2  
@Jack: I wonder how many times you will ask/erase/re-ask somewhere else the same (trivial) question. This is an odd way of showing your appreciation of the answers you receive. –  Did Jun 10 '11 at 14:37

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.