Log-domain interior-point methods for convex quadratic programming

被引:5
作者
Permenter, Frank [1 ]
机构
[1] Toyota Res Inst, Cambridge, MA 02139 USA
关键词
Quadratic programming; Convex optimization; Interior-point methods; POTENTIAL REDUCTION ALGORITHM; PATH;
D O I
10.1007/s11590-022-01952-z
中图分类号
C93 [管理学]; O22 [运筹学];
学科分类号
070105 ; 12 ; 1201 ; 1202 ; 120202 ;
摘要
Applying an interior-point method to the central-path conditions is a widely used approach for solving quadratic programs. Reformulating these conditions in the log-domain is a natural variation on this approach that to our knowledge is previously unstudied. In this paper, we analyze log-domain interior-point methods and prove their polynomial-time convergence. We also prove that they are approximated by classical barrier methods in a precise sense and provide simple computational experiments illustrating their superior performance.
引用
收藏
页码:1613 / 1631
页数:19
相关论文
共 19 条
[1]  
Achache M, 2006, COMPUT APPL MATH, V25, P97
[2]   On implementing a primal-dual interior-point method for conic quadratic optimization [J].
Andersen, ED ;
Roos, C ;
Terlaky, T .
MATHEMATICAL PROGRAMMING, 2003, 95 (02) :249-277
[3]  
[Anonymous], 2013, Interior point methods of mathematical programming
[4]  
Arora S., 2012, Theory of Computing, V8, P121, DOI 10.4086/toc.2012.v008a006
[5]  
Di Cairano S, 2013, IEEE DECIS CONTR P, P1696, DOI 10.1109/CDC.2013.6760126
[6]  
Dorn WS., 1960, Q APPL MATH, V18, P155, DOI DOI 10.1090/QAM/112751
[7]  
Drummond LMG, 2000, RAIRO-RECH OPER, V34, P331, DOI 10.1051/ro:2000117
[8]   AN O(N(3)L) PRIMAL DUAL POTENTIAL REDUCTION ALGORITHM FOR SOLVING CONVEX QUADRATIC PROGRAMS [J].
GOLDFARB, D ;
LIU, SC .
MATHEMATICAL PROGRAMMING, 1993, 61 (02) :161-170
[9]  
GOLDFARB D, 1991, MATH PROGRAM, V49, P325
[10]  
Kapoor S., 1986, P 18 ANN ACM S THEOR, P147, DOI DOI 10.1145/12130.12145