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.

Let $C$ be the set of smooth curves from the unit interval into $\mathbb{R}^n$. Let $f : C \rightarrow \mathbb{R}$ be a functional on these curves given by $f(x) := \int_0^1 L(x,\dot{x}) dt$. Define the Hessian of $f$ at $x(t)$ as $f_{**}(V,W)$ which a bi-linear form on vector fields along $x(t)$. This is defined by first taking a variation of curves $x(h_1, h_2)$ satisfying $x(0,0) = x$ and $\frac{\partial x(h_1, h_2)}{\partial h_1}|_{(0,0)} = V$ and $\frac{\partial x(h_1, h_2)}{\partial h_2}|_{(0,0)} = W$. Then we set $f_{**}(V,W) = \frac{\partial^2 f(x(h_1, h_2))}{\partial h_1 \partial h_2}|_{(0,0)}$

What conditions would I need on $f$ to show the Hessian of $f$ is positive or negative definite. ie) Are there any standard techniques to show $f$ is either a minimum or a maximum at $x$. I've seen a theorem for geodesics which basically says that if there are no conjugate points along $x(t)$, then we are guaranteed the geodesic will be a local minimum but the proof for this usually involves geometrical argument specifically for this functional. Does such a theorem generalise for the equivalent notion of conjugacy for $f$ given we place suitable constraints on $f$?

Anything to enlighten me on this topic would be appreciated. Thanks.

share|improve this question
add comment

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.