Efficient Implementations of Interior-Point Methods for Integral Quadratic Constraints

• We describe two strategies for exploiting structure in implementations of interior-point methods for the semidefinite programs (SDPs) that result from integral quadratic constraints. The first approach uses inexact search directions computed by the conjugate gradient algorithm. In the second approach we solve the problem via the dual. Simplifying the dual problem results in an SDP that has fewer variables (O(n)) than the primal SDP (O(n2) variables), and that can be solved efficiently using standard methods.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Reglerteknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Control Engineering (hsv//eng)

Nyckelord

Integral quadratic constraints

Semidefinite programs

TECHNOLOGY

TEKNIKVETENSKAP

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)