Virtuous smoothing for global optimization

Jon Lee, Daphne Skipper

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In the context of global optimization and mixed-integer non-linear programming, generalizing a technique of D’Ambrosio, Fampa, Lee and Vigerske for handling the square-root function, we develop a virtuous smoothing method, using cubics, aimed at functions having some limited non-smoothness. Our results pertain to root functions (wp with 0 < p< 1) and their increasing concave relatives. We provide (i) a sufficient condition (which applies to functions more general than root functions) for our smoothing to be increasing and concave, (ii) a proof that when p= 1 / q for integers q≥ 2 , our smoothing lower bounds the root function, (iii) substantial progress (i.e., a proof for integers 2 ≤ q≤ 10 , 000) on the conjecture that our smoothing is a sharper bound on the root function than the natural and simpler “shifted root function”, and (iv) for all root functions, a quantification of the superiority (in an average sense) of our smoothing versus the shifted root function near 0.

Original languageEnglish (US)
Pages (from-to)677-697
Number of pages21
JournalJournal of Global Optimization
Volume69
Issue number3
DOIs
StatePublished - Nov 1 2017

Fingerprint

Global optimization
Global Optimization
Smoothing
Roots
Q-integers
Mixed Integer Nonlinear Programming
Smoothing Methods
Sharp Bound
Nonlinear programming
Square root
Quantification
Integer
Lower bound
Sufficient Conditions

Keywords

  • Global optimization
  • Non-differentiable
  • Non-smooth
  • Piece-wise functions
  • Roots
  • Smoothing

ASJC Scopus subject areas

  • Computer Science Applications
  • Control and Optimization
  • Management Science and Operations Research
  • Applied Mathematics

Cite this

Virtuous smoothing for global optimization. / Lee, Jon; Skipper, Daphne.

In: Journal of Global Optimization, Vol. 69, No. 3, 01.11.2017, p. 677-697.

Research output: Contribution to journalArticle

Lee, Jon ; Skipper, Daphne. / Virtuous smoothing for global optimization. In: Journal of Global Optimization. 2017 ; Vol. 69, No. 3. pp. 677-697.
@article{e3e1d3cc62fe4aa3a9f0d6c5b83abdbe,
title = "Virtuous smoothing for global optimization",
abstract = "In the context of global optimization and mixed-integer non-linear programming, generalizing a technique of D’Ambrosio, Fampa, Lee and Vigerske for handling the square-root function, we develop a virtuous smoothing method, using cubics, aimed at functions having some limited non-smoothness. Our results pertain to root functions (wp with 0 < p< 1) and their increasing concave relatives. We provide (i) a sufficient condition (which applies to functions more general than root functions) for our smoothing to be increasing and concave, (ii) a proof that when p= 1 / q for integers q≥ 2 , our smoothing lower bounds the root function, (iii) substantial progress (i.e., a proof for integers 2 ≤ q≤ 10 , 000) on the conjecture that our smoothing is a sharper bound on the root function than the natural and simpler “shifted root function”, and (iv) for all root functions, a quantification of the superiority (in an average sense) of our smoothing versus the shifted root function near 0.",
keywords = "Global optimization, Non-differentiable, Non-smooth, Piece-wise functions, Roots, Smoothing",
author = "Jon Lee and Daphne Skipper",
year = "2017",
month = "11",
day = "1",
doi = "10.1007/s10898-017-0533-x",
language = "English (US)",
volume = "69",
pages = "677--697",
journal = "Journal of Global Optimization",
issn = "0925-5001",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Virtuous smoothing for global optimization

AU - Lee, Jon

AU - Skipper, Daphne

PY - 2017/11/1

Y1 - 2017/11/1

N2 - In the context of global optimization and mixed-integer non-linear programming, generalizing a technique of D’Ambrosio, Fampa, Lee and Vigerske for handling the square-root function, we develop a virtuous smoothing method, using cubics, aimed at functions having some limited non-smoothness. Our results pertain to root functions (wp with 0 < p< 1) and their increasing concave relatives. We provide (i) a sufficient condition (which applies to functions more general than root functions) for our smoothing to be increasing and concave, (ii) a proof that when p= 1 / q for integers q≥ 2 , our smoothing lower bounds the root function, (iii) substantial progress (i.e., a proof for integers 2 ≤ q≤ 10 , 000) on the conjecture that our smoothing is a sharper bound on the root function than the natural and simpler “shifted root function”, and (iv) for all root functions, a quantification of the superiority (in an average sense) of our smoothing versus the shifted root function near 0.

AB - In the context of global optimization and mixed-integer non-linear programming, generalizing a technique of D’Ambrosio, Fampa, Lee and Vigerske for handling the square-root function, we develop a virtuous smoothing method, using cubics, aimed at functions having some limited non-smoothness. Our results pertain to root functions (wp with 0 < p< 1) and their increasing concave relatives. We provide (i) a sufficient condition (which applies to functions more general than root functions) for our smoothing to be increasing and concave, (ii) a proof that when p= 1 / q for integers q≥ 2 , our smoothing lower bounds the root function, (iii) substantial progress (i.e., a proof for integers 2 ≤ q≤ 10 , 000) on the conjecture that our smoothing is a sharper bound on the root function than the natural and simpler “shifted root function”, and (iv) for all root functions, a quantification of the superiority (in an average sense) of our smoothing versus the shifted root function near 0.

KW - Global optimization

KW - Non-differentiable

KW - Non-smooth

KW - Piece-wise functions

KW - Roots

KW - Smoothing

UR - http://www.scopus.com/inward/record.url?scp=85019708260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019708260&partnerID=8YFLogxK

U2 - 10.1007/s10898-017-0533-x

DO - 10.1007/s10898-017-0533-x

M3 - Article

AN - SCOPUS:85019708260

VL - 69

SP - 677

EP - 697

JO - Journal of Global Optimization

JF - Journal of Global Optimization

SN - 0925-5001

IS - 3

ER -