The present project aims at studying and implementing algorithms for deterministic global optimization of nonlinear programming problems. The research will be focused on problems with continuous second order derivatives and make use of modern techniques of computational differentiation (automatic differentiation). It is also intended to study optimization problems where the functions involved are grey box (i.e., functions that do not have derivatives available, or that do not have analytical expression available, or that come from simulations, or functions whose cost of evaluation is high). The main goal of the project will be the development of a robust and efficient software for global optimization.
News published in Agência FAPESP Newsletter about the scholarship: