Skip to main navigation Skip to search Skip to main content

On the Convexification of Non-Linear Optimization Problems under Performance Guarantees

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The convexification operator is the mapping that determines for every given continuous function on the real axis its greatest convex minorant. It is needed to convexify non-convex optimization problems in order to apply powerful convex optimization techniques to solve them. This paper shows that the convexification operator cannot be implemented on a Turing machine. In particular, we show that there exist piecewise linear continuous functions with a unique global minimum such that their greatest convex minorant is not Turing computable. Furthermore, for these functions there do not even exist monotonically increasing, computable sequences of convex computable continuous functions that converge to the greatest convex minorant. The same results also hold for the least concave majorants of continuous functions.

Original languageEnglish (US)
Title of host publication2025 IEEE 64th Conference on Decision and Control, CDC 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages8136-8142
Number of pages7
ISBN (Electronic)9798331526276
DOIs
StatePublished - 2025
Externally publishedYes
Event64th IEEE Conference on Decision and Control, CDC 2025 - Rio de Janeiro, Brazil
Duration: Dec 9 2025Dec 12 2025

Publication series

NameProceedings of the IEEE Conference on Decision and Control
ISSN (Print)0743-1546
ISSN (Electronic)2576-2370

Conference

Conference64th IEEE Conference on Decision and Control, CDC 2025
Country/TerritoryBrazil
CityRio de Janeiro
Period12/9/2512/12/25

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Fingerprint

Dive into the research topics of 'On the Convexification of Non-Linear Optimization Problems under Performance Guarantees'. Together they form a unique fingerprint.

Cite this