LOQO is a powerful solver for smooth constrained optimization problems, based on interior-point method applied to a sequence of quadratic approximations. Subject to the requirement that the defining functions be smooth (at the points evaluated by the algorithm), LOQO can handle a range of problems: linear or nonlinear, convex or nonconvex, constrained or unconstrained. For convex problems, LOQO finds a globally optimal solution; otherwise, it iterates from the given starting point to find a locally optimal solution.
Developer: Prof. Robert Vanderbei, Princeton University
Current version: 7.03
Problem types supported: Linear, quadratic, and smooth nonlinear objectives and constraints in continuous variables.
Algorithms available: An infeasible-start, primal-dual interior-point method for linear and quadratic programming, with robust extensions to more general nonlinear objective and constraint functions.
Special features: Option to assert that problem is convex, for greater efficiency.
LOQO User’s Manual
LOQO for AMPL option listing