# Inverse Function Theorem for functions $f(x,y)$ and $\int\limits_0^1\frac{\partial f}{\partial x}(tx,y)dt$

I'm struggling with the following problem:

Let $$f\colon\mathbb{R}^2\to\mathbb{R}$$ be a twice continuously differentiable function satisfying $$f(0,y)=0\mbox{ for all }y\in\mathbb{R}$$

(a) Show that $$f(x,y) = xg(x,y)$$ for all pairs $$(x,y)\in\mathbb{R}^2$$, where $$g$$ is the function given by $$g(x,y) = \int\limits_0^1\frac{\partial f}{\partial x}(tx,y)dt$$

(b) Show that $$g$$ is continuously differentiable and that for all $$x\in\mathbb{R}$$ $$g(0,y) = \frac{\partial f}{\partial x}(0,y),~\frac{\partial g}{\partial y}(0,y) = \frac{\partial^2 f}{\partial x\partial y}(0,y)$$

(c) If $$\frac{\partial f}{\partial x}(0,0)\neq0$$ there is a neighborhood $$V$$ of $$(0,0)$$ in $$\mathbb{R}^2$$ such that $$f^{-1}(0)\cap V = V\cap \{x=0\}$$

(d) If $$\frac{\partial f}{\partial x}(0,0) = 0$$ and $$\frac{\partial^2 f}{\partial x\partial y}(0,0)\neq0$$ there is a neighborhood $$V$$ of $$(0,0)$$ in $$\mathbb{R}^2$$ such that $$f^{-1}(0)\cap V$$ consists of the union of the set $$V\cap\{x=0\}$$ with a curve through $$(0,0)$$, whose tangent at $$(0,0)$$ is not vertical (not parallel to the $$y$$-axis)

Here are my attempts:

(a) Note that $$\frac{\partial f}{\partial x}(tx,y) = \frac{t}{x}\cdot\frac{\partial f}{\partial t}(tx,y)$$ and so using the integration by parts $$g(x,y) = \int\limits_0^1\frac{\partial f}{\partial x}(tx,y)dt = \int\limits_0^1 \frac{t}{x}\cdot\frac{\partial f}{\partial t}(tx,y)dt = \frac{1}{x}\left(f(x,y) - \int\limits_0^1 f(tx,y)dt\right)$$ it suffices to show that $$\int\limits_0^1 f(tx,y)dt = 0$$, however, I can't see how to proceed with it.

(b) Expressions for $$g(0,y)$$ and $$\frac{\partial g}{\partial y}(0,y)$$ is the result of a straightforward calculation using the definiton of $$g(x,y)$$, but I have troubles showing that $$g$$ is continuously differentiable. What am I supposed to do -- prove that the partial derivatives exist and they are continuous? If so, can someone write down the details?

(c) One can fix $$y=0$$ and consider a function $$f(x,0)\colon\mathbb{R}\to\mathbb{R}$$. Then, since $$\frac{\partial f}{\partial x}(0,0)\neq0$$, the Inverse Function Theorem says that there exists a neighborhood $$U$$ of $$0\in\mathbb{R}$$ such that $$f\colon U\to f(U)$$ is the bijection. Denote $$V = U\times\{0\}$$, so $$V\subset\mathbb{R}^2$$ and we are done. Am I right here?

(d) It seems like using the result of part (b) we can do the same trick with the function $$g(0,y)\colon\mathbb{R}\to\mathbb{R}$$, but I can't finish the proof.

Any help will be really appreciated. Thanks a lot in advance.

• Can you make the title somehow informative and descriptive, rather than subjective? (Do describe the equation; Don't say "tricky problem".) – Asaf Karagila May 17 at 11:10
• @AsafKaragila, I did it. – Hasek May 17 at 17:52

I think you are misunderstanding something with the integral. $$\frac{\partial f}{\partial x}(tx,y)$$ doesn't mean the derivative with respect to the $$x$$ inside, it means the derivative of the function with respect to the first argument. In other words the integral really evaluates to
$$g(x,y) = \int_0^1 \frac{\partial f}{\partial x}(tx,y)\:dt = \begin{cases} \frac{1}{x}f(tx,y)\Bigr|_0^1 = \frac{f(x,y)-f(0,y)}{x} = \frac{1}{x}f(x,y) & x\neq0 \\ \int_0^1 \frac{\partial f}{\partial x}(0,y) \:dt = \frac{\partial f}{\partial x}(0,y) & x= 0 \\ \end{cases}$$
therefore we have in both cases that $$f(x,y) = xg(x,y)$$.
In order to show $$g$$ is continuous differentiable it is enough (sufficient but not necessary) to show that the partial derivatives of $$g$$ exist and are continuous.
The other questions are just application of the inverse function theorem on $$g$$ and its first partial, which is why the question has you prove differentiability first. I believe your proof is valid for part (c), so just continue on with (d).
• Part (d) is still unclear for me. If I will consider $g(0,y)\colon\mathbb{R}\to\mathbb{R}$, then, since $\frac{\partial g}{\partial y}(0,0)\neq0$, from the Inverse Function Theorem for $g(0,y)$ there exist a neighborhood $W\subset\mathbb{R}$ such that $g\colon W\to g(W)$ is a bijection, i.e. $g(0,0)=0$ will have exactly one $y=0$ as a preimage. This is where I'm lost. I can't link it with the preimage of $f$ (not $g$!) and I can`t understand where should this curve with a non-vertical tangent come from. Can you please elaborate on it? – Hasek May 17 at 21:24