An algorithm for computing solutions of variational problems with global convexity constraints (Q2270139)

The authors present an algorithm to approximate the solutions to variational problems of the form \[ \underset{v\in C}{}{\text{inf}}\,|[v]|= \int_\Theta L(\theta, v(\theta),\nabla v(\theta)) f(\theta)\,d\theta, \] where \(L(\theta,z,p)= z- \theta\cdot p+ C(p)\), \(f\) is a probability density on \(\Theta\) and the set of admissible functions consists of convex functions. The motivation behind this numerical algorithm is to compute solutions to adverse selection problems within a principal-agent framework.