On strong duality, theorems of the alternative, and projections in conic optimization

Abstract

A conic program is the problem of optimizing a linear function over a closed convex cone intersected with an affine preimage of another cone. We analyse three constraint qualifications, namely a Closedness CQ, Slater CQ, and Boundedness CQ (also called Clark-Duffin theorem), that are sufficient for achieving strong duality and show that the first implies the second which implies the third, and also give a more general form of the third CQ for conic problems. Furthermore, two consequences of strong duality are presented, the first being a theorem of the alternative on almost feasibility (also called weak infeasibility), and the second being an explicit description of the projection of conic sets onto linear subspaces, akin to using projection cones for polyhedral sets.

arXiv commentRevised version 2 by adding a more thorough literature review. In particular, the authors were made aware that a similar proof for Closedness CQ given in previous versions has appeared in Barvinok's book, and that there are some related results on conic duality in the unpublished technical report of Luo et al