I've proven that the function $f: U=(0,\infty)\times \mathbb R \rightarrow \mathbb R^2$ given by $f(x,y) = (x, y^3 + xy)$ is injective and surjective ($f(U) = U$), so it is bijective.

I've computed $\frac {\partial f_1}{\partial x} = 1$, $\frac {\partial f_1}{\partial y} = 0$, $\frac {\partial f_2}{\partial x} = y$ and $\frac {\partial f_2}{\partial y} = 3y^2+x$.

Can someone tell me how to apply The Inverse Function Theorem to prove $f$ is a diffeomorphism ? ($f$ is smooth is easy to see, I've shown it is bijective). I need only to verify it has a smooth inverse function $f^{-1}: U \rightarrow U$.

I find the theorem confusing to apply.


You compute the determinant of the Jacobian matrix (of which you computed already its components). The result is $1\times(2y^2+x)-0\times y=2y^2+x$. Then you can see that this is never zero on $U$ ($x>0$ and $y^2\geq0$).

The inverse function theorem gives you the differentiability of the inverse.

  • $\begingroup$ So this function $g$ one gets from TIFT must be equal to $f^{-1}$ by the property that $f(g(v))=v$ for all $v$ in the small open region in $f(U)$ ? And since $g$ is smooth, this implies $f^{-1}$ is smooth ? $\endgroup$ – Shuzheng Feb 21 '15 at 18:10
  • $\begingroup$ Yes, if $f$ has an inverse and $f(g(v))=v$ then $g$ is that inverse. $\endgroup$ – user218170 Feb 21 '15 at 18:13
  • $\begingroup$ Ahh, and since the region is open around this function $g$, it satisfies the definition of taking the derivative :) Thanks ! $\endgroup$ – Shuzheng Feb 21 '15 at 18:21

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.