A New Class of Polynomial Interior-Point Algorithms for P*(κ)-Linear Complementarity Problems

Goran Lesaja

A New Class of Polynomial Interior-Point Algorithms for P*(κ)-Linear Complementarity Problems

Mathematical Sciences

Research output: Contribution to conference › Presentation

Abstract

We present a new class of short- and long-step polynomial interior-point algorithms for P*(κ)- Linear Complementarity Problems based on a new class of parametric kernel functions. This class is fairly general and includes the classical logarithmic function, the prototype self-regular function, and non-self-regular kernel functions as special cases. The iteration bounds obtained in this paper are very favorable and they match the best known existing iteration bounds

Original language	American English
State	Published - Aug 13 2007
Event	International Conference on Continuous Optimization (ICCOPT) - Duration: Aug 13 2007 → …

Conference

Conference	International Conference on Continuous Optimization (ICCOPT)
Period	08/13/07 → …

Keywords

Algorithms
Interior-Point
Linear Complementarity Problems
Polynomial

DC Disciplines

Mathematics
Physical Sciences and Mathematics

Cite this

@conference{169cef310ae94604bbcea377471f45fb,

title = "A New Class of Polynomial Interior-Point Algorithms for P*(κ)-Linear Complementarity Problems",

abstract = " We present a new class of short- and long-step polynomial interior-point algorithms for P*(κ)- Linear Complementarity Problems based on a new class of parametric kernel functions. This class is fairly general and includes the classical logarithmic function, the prototype self-regular function, and non-self-regular kernel functions as special cases. The iteration bounds obtained in this paper are very favorable and they match the best known existing iteration bounds",

keywords = "Algorithms, Interior-Point, Linear Complementarity Problems, Polynomial",

author = "Goran Lesaja",

year = "2007",

month = aug,

day = "13",

language = "American English",

note = "International Conference on Continuous Optimization (ICCOPT) ; Conference date: 13-08-2007",

}

TY - CONF

T1 - A New Class of Polynomial Interior-Point Algorithms for P*(κ)-Linear Complementarity Problems

AU - Lesaja, Goran

PY - 2007/8/13

Y1 - 2007/8/13

N2 - We present a new class of short- and long-step polynomial interior-point algorithms for P*(κ)- Linear Complementarity Problems based on a new class of parametric kernel functions. This class is fairly general and includes the classical logarithmic function, the prototype self-regular function, and non-self-regular kernel functions as special cases. The iteration bounds obtained in this paper are very favorable and they match the best known existing iteration bounds

AB - We present a new class of short- and long-step polynomial interior-point algorithms for P*(κ)- Linear Complementarity Problems based on a new class of parametric kernel functions. This class is fairly general and includes the classical logarithmic function, the prototype self-regular function, and non-self-regular kernel functions as special cases. The iteration bounds obtained in this paper are very favorable and they match the best known existing iteration bounds

KW - Algorithms

KW - Interior-Point

KW - Linear Complementarity Problems

KW - Polynomial

UR - https://digitalcommons.georgiasouthern.edu/math-sci-facpres/304

M3 - Presentation

T2 - International Conference on Continuous Optimization (ICCOPT)

Y2 - 13 August 2007

ER -

A New Class of Polynomial Interior-Point Algorithms for P*(κ)-Linear Complementarity Problems

Abstract

Conference

Keywords

DC Disciplines

Other files and links

Fingerprint

Cite this