A cut-based algorithm for the nonlinear dual of the minimum cost network flow problem

被引:20
作者
Ahuja, RK [1 ]
Hochbaum, DS
Orlin, JB
机构
[1] Univ Florida, Gainesville, FL 32611 USA
[2] Univ Calif Berkeley, Ind Engn & Operat Res, Berkeley, CA 94720 USA
[3] Univ Calif Berkeley, Walter A Haas Sch Business, Berkeley, CA 94720 USA
[4] MIT, Sloan Sch Management, Cambridge, MA 02139 USA
关键词
nonlinear integer programming; convex integer programming; total unimodularity; minimum cut; network flow;
D O I
10.1007/s00453-004-1085-2
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
We consider a convex, or nonlinear. separable minimization problem with constraints that are 44 dual to the minimum cost network flow problem. We show how to reduce this problem to a polynomial number of minimum s.t-cut problems. The solution of the reduced problem utilizes the technique for solving integer programs on monotone inequalities in three variables, and a so-called proximity-scaling technique that reduces a convex problem to its linear objective counterpart. The problem is solved in this case in a logarithmic number 14 of calls. O(log U). to a minimum cut procedure, where U is the range of the variables. For a convex problem on n variables the minimum cut is solved on a graph with O(n(2)) nodes. Among the consequences of this result is a new cut-based scaling algorithm for the minimum cost network flow problem. When the objective function is an arbitrary nonlinear function we demonstrate that this constrained problem is solved in pseudopolynomial time by applying a minimum cut procedure to a graph on O(nU) nodes.
引用
收藏
页码:189 / 208
页数:20
相关论文
共 21 条
[11]   CONVEX SEPARABLE OPTIMIZATION IS NOT MUCH HARDER THAN LINEAR OPTIMIZATION [J].
HOCHBAUM, DS ;
SHANTHIKUMAR, JG .
JOURNAL OF THE ACM, 1990, 37 (04) :843-862
[12]   Minimizing a convex cost closure set [J].
Hochbaum, DS ;
Queyranne, M .
SIAM JOURNAL ON DISCRETE MATHEMATICS, 2003, 16 (02) :192-207
[13]   Solving integer programs over monotone inequalities in three variables: A framework for half integrality and good approximations [J].
Hochbaum, DS .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2002, 140 (02) :291-321
[14]   An efficient algorithm for image segmentation, Markov random fields and related problems [J].
Hochbaum, DS .
JOURNAL OF THE ACM, 2001, 48 (04) :686-701
[15]   TIGHT BOUNDS AND 2-APPROXIMATION ALGORITHMS FOR INTEGER PROGRAMS WITH 2 VARIABLES PER INEQUALITY [J].
HOCHBAUM, DS ;
MEGIDDO, N ;
NAOR, J ;
TAMIR, A .
MATHEMATICAL PROGRAMMING, 1993, 62 (01) :69-83
[16]   Efficient algorithms for the inverse spanning-tree problem [J].
Hochbaum, DS .
OPERATIONS RESEARCH, 2003, 51 (05) :785-797
[17]  
HOCHBAUM DS, 1998, PSEUDOFLOW AGORITHMS
[18]  
HOCHBAUM DS, 2001, INVERSE SHORTEST PAT
[19]   Polynomial methods for separable convex optimization in unimodular linear spaces with applications [J].
Karzanov, AV ;
McCormick, ST .
SIAM JOURNAL ON COMPUTING, 1997, 26 (04) :1245-1275
[20]   MAXIMAL CLOSURE OF A GRAPH AND APPLICATIONS TO COMBINATORIAL PROBLEMS [J].
PICARD, JC .
MANAGEMENT SCIENCE, 1976, 22 (11) :1268-1272