Legendre transform and applications to finite and infinite optimization (Q505632): Difference between revisions

The paper investigates convex constrained nonlinear optimization problems and optimal control with convex constraints in the light of the so-called Legendre transform. This change of coordinates is used to propose a gradient-like algorithm for mathematical programs, which can be seen as a search method along geodesics. The Legendre transform is also used to study the value function of a state constrained Mayer problem.