I am preparing for an exam and I cannot seem to figure out how to solve exercises where I need to apply the implicit function theorem.


Let $h: \mathbb{R}^2 \rightarrow \mathbb{R}$ given by

$\large h(u,v) = u^2 + (v-1)^2 - 5 + e^{u-2}$

$\large(i)$ Show that $h(2,1) = 0$ and $h\in C^1(\mathbb{R}^2)$

$h(2,1) = 2^2 + (1-1)^2 - 5 + e^{2-2} = 4+0-5+1 = 0.$

To show that $\textbf{h}\in C^1(\mathbb{R}^2)$ I show that the partial derivatives exists and that they are continous. I have not included the calculations but the partial derivatives are

\begin{align*} \large h_u(u, v) = & 2u + e^{(u-2)} \end{align*}

\begin{align*} \large h_v(u, v)=&2(v-1) \end{align*}

Now the next exercise is the one I am having problems with..

$\large (ii)$ Show that one can apply the implicit function theorem in order to obtain some small enough $\epsilon > 0$ and a $C^1$ function $f: (1-\epsilon, 1 + \epsilon) \rightarrow \mathbb{R}$ such that

$$\large h(f(v),v)=0, ~~~~\forall v \in (1-\epsilon, 1 + \epsilon) $$

MY APPROACH: The Implicit function theorem states that if $(x^*,y^*) \in \mathbb{R}^2$ is a particular solution of $G(x^*,y^*) = c$ and $\dfrac{\delta G}{\delta y}(x^*,y^*) \neq 0$, then the equation

$$ G(x,y)=c$$ determines a function $y=y(x)$ defined on some interval I$=(x^*-\epsilon, x^*+\epsilon)$ about the point $x^*$ such that

(a) $G(x, y(x))=c$ for all $x \in $I

(b) $y(x^*)=y^*$

(c) $y'(x^*)= -\frac{\dfrac{\delta G}{\delta x}(x^*,y^*)}{\dfrac{\delta G}{\delta y}(x^*,y^*)}$

I have that $h(2,1)=0$ thus for $h(f(v),v)=0$ I need to have $f(1)=2$.

I have that \begin{align*} \large h_u(2, 1) = & 4 + 1 = 5 \neq 0 \end{align*}

but I have that

\begin{align*} \large h_v(2, 1) = 2(1-1) = 0 \end{align*}

so I get that $\large f'(1) = \frac{\frac{\delta h}{\delta v}(2,1)}{\frac{\delta h}{\delta u}(2,1)} = \frac{0}{5} = 0 $

How do I find the function $f(v)$? I cannot see how I proceed from here. My first thought was linear approximation, but I am not sure if it is correct.

Best Husky


You're not requested to find the function $f$ but to prove that it exists. That makes a big difference!

Also notice that the implicit function theorem that you mention is related to the second variable while your exercise is related to the first one.

  • $\begingroup$ Ah I see, I have totally misunderstood it then. I saw "show" as "find". haha. Is the above enough to prove that it exists? And yes I noticed that it is related to the first variable, otherwise I would have to divide with zero. $\endgroup$ – Husky653 Jun 10 '15 at 14:46
  • 1
    $\begingroup$ Yes it is sufficient. $\endgroup$ – mathcounterexamples.net Jun 10 '15 at 14:55
  • $\begingroup$ In (iii) I am asked to find f'(1), this is just the result from (ii) right? That is f'(1)=0. $\endgroup$ – Husky653 Jun 10 '15 at 15:03
  • 1
    $\begingroup$ That's the way ro proceed. $\endgroup$ – mathcounterexamples.net Jun 10 '15 at 15:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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