Abstract:

Recent advances in the efficiency and robustness of algorithms solving convex quadratically constrained quadratic programming (QCQP) problems motivate developing techniques for cre- ating convex quadratic relaxations that, although more expensive to compute, provide tighter bounds than their classical linear counterparts. In the first part of this two-paper series [Strahl et al., 2024], we developed a cutting-plane algorithm to construct convex quadratic underestimators for twice-differentiable convex functions, which we extend here to address the case of non-convex difference-of-convex (d.c.) functions as well. Furthermore, we generalize our approach to consider a hierarchy of quadratic forms, thereby allowing the construction of even tighter underestimators. Utilizing a benchmark library of d.c. functions, we demonstrate note- worthy reduction in the hypervolume between our quadratic underestimators and linear ones constructed at the same points. Additionally, we construct convex QCQP relaxations at the root node of a spatial branch-and-bound tree for a set of systematically created d.c. optimization problems in up to four dimensions, and we show that our relaxations reduce the gap between the lower bound computed by the state-of-the-art global optimization solver BARON and the optimal solution by an excess of 90%, on average.