Abstract:Deterministic nonconvex optimization solvers generate convex relaxations of nonconvex functions by making use of underlying factorable representations. One approach introduces auxiliary variables assigned to each factor that lifts the problem into a higher-dimensional decision space. In contrast, a generalized McCormick relaxation approach offers the significant advantage of constructing relaxations in the lower dimensionality space of the original problem without introducing auxiliary variables, often referred to as a "reduced-space" approach. Recent contributions illustrated how additional nontrivial inequality constraints may be used in factorable programming to tighten relaxations of the ubiquitous bilinear term. In this work, we exploit an analogous representation of McCormick relaxations and factorable programming to formulate tighter relaxations in the original decision space. We develop the underlying theory to generate necessarily tighter reduced-space McCormick relaxations when a priori convex/concave relaxations are known for intermediate bilinear terms. We then show how these rules can be generalized within a McCormick relaxation scheme via three different approaches: the use of a McCormick relaxations coupled to affine arithmetic, the propagation of affine relaxations implied by subgradients, and an enumerative approach that directly uses relaxations of each factor. The developed approaches are benchmarked on a library of optimization problems using the EAGO.jl optimizer. Two case studies are also considered to demonstrate the developments: an application in advanced manufacturing to optimize supply chain quality metrics and a global dynamic optimization application for rigorous model validation of a kinetic mechanism. The presented subgradient method leads to an improvement in CPU time required to solve the considered problems to -global optimality.

Relaxation strength for multilinear optimization: McCormick strikes back

On the strength of recursive McCormick relaxations for binary polynomial optimization

Linearization of McCormick relaxations and hybridization with the auxiliary variable method

Improved Convex and Concave Relaxations of Composite Bilinear Forms

Recursive McCormick Linearization of Multilinear Programs

McCormick envelopes in mixed-integer PDE-constrained optimization

MIP Relaxations in Factorable Programming

Piecewise Polyhedral Relaxations of Multilinear Optimization

On the strength of Burer's lifted convex relaxation to quadratic programming with ball constraints

Relaxations and cutting planes for linear programs with complementarity constraints

A Correlatively Sparse Lagrange Multiplier Expression Relaxation for Polynomial Optimization

On Tractable Convex Relaxations of Standard Quadratic Optimization Problems under Sparsity Constraints

Complexity of linear relaxations in integer programming

Instance-specific linear relaxations of semidefinite optimization problems

Piecewise Polyhedral Formulations for a Multilinear Term

Efficient treatment of bilinear forms in global optimization

Strong underrelaxation in Kaczmarz's method for inconsistent systems

Convexification of Bilinear Terms over Network Polytopes

Occupation Measure Relaxations in Variational Problems: The Role of Convexity