Primal-dual interior-point methods for second-order conic optimization based on self-regular proximities

被引:46
作者
Peng, JM [1 ]
Roos, C
Terlaky, T
机构
[1] McMaster Univ, Dept Comp & Software, Hamilton, ON L8S 4L7, Canada
[2] Delft Univ Technol, Fac Informat Technol & Syst, NL-2600 GA Delft, Netherlands
关键词
second-order conic optimization; primal-dual interior-point method; self-regular proximity function; polynomial complexity;
D O I
10.1137/S1052623401383236
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Recently the authors introduced the notions of self-regular functions and self-regular proximity functions and used them in the design and analysis of interior-point methods (IPMs) for linear and semidefinite optimization (LO and SDO). In this paper, we consider an extension of these concepts to second-order conic optimization (SOCO). This nontrivial extension requires the development of various new tools. Versatile properties of general analytical functions associated with the second-order cone are exploited. Based on the so-called self-regular proximity functions, new primal-dual Newton methods for solving SOCO problems are proposed. It will be shown that these new large-update IPMs for SOCO enjoy polynomial O(max{p, q}N(q+1)/2q logN/epsilon) iteration bounds analogous to those of their LO and SDO cousins, where N is the number of constraining cones and p, q are constants, the so-called growth degree and barrier degree of the corresponding proximity. Our analysis allows us to choose not only a constant q but also a q as large as log N. In this case, our new algorithm has the best known O(N-1/2 log N logN/epsilon) iteration bound for large-update IPMs.
引用
收藏
页码:179 / 203
页数:25
相关论文
共 27 条
[21]  
SCHMIETA S, 1999, 1399 RRR RUTG U RUTG
[22]   Associative and Jordan algebras, and polynomial time interior-point algorithms for symmetric cones [J].
Schmieta, SH ;
Alizadeh, F .
MATHEMATICS OF OPERATIONS RESEARCH, 2001, 26 (03) :543-564
[23]  
STURDY J, 1999, HON DEF AV SYST 13 I, P3
[24]   A convergence analysis of the scaling-invariant primal-dual path-following algorithms for second-order cone programming [J].
Tsuchiya, T .
OPTIMIZATION METHODS & SOFTWARE, 1999, 11-2 (1-4) :141-182
[25]  
TSUCHIYA T, 1997, 649 I STAT MATH
[26]  
WOLKOWICZ H, 2000, HDB SEMIDEFINITE PRO
[27]  
[No title captured]