Affiliation: | (1) Department of Mathematical Engineering and Information Physics, University of Tokyo, Tokyo, 113-8656, Japan;(2) Faculty of Commerce and Business Administration, University of British Columbia, Vancouver, BC V6T 1Z2, Canada;(3) Institute of Policy and Planning Sciences, University of Tsukuba, Tsukuba, Ibaraki, 305, Japan |
Abstract: | This paper presents two fast cycle canceling algorithmsfor the submodular ow problem. The rst uses an assignmentproblem whose optimal solution identies most negativenode-disjoint cycles in an auxiliary network. Canceling thesecycles lexicographically makes it possible to obtain an optimalsubmodular ow in O(n 4 h log(nC)) time, which almost matches thecurrent fastest weakly polynomial time for submodular flow(where n is the number ofnodes, h is the time forcomputing an exchange capacity, and C is the maximum absolute value of arccosts). The second algorithm generalizes Goldbergs cyclecanceling algorithm for min cost flow to submodular flow to alsoget a running time of O(n 4 h log(nC)).. We show how to modify thesealgorithms to make them strongly polynomial, with running timesof O(n 6 h log n), which matches the fastest stronglypolynomial time bound for submodular flow. We also show how toextend both algorithms to solve submodular flow with separableconvex objectives. * An extended abstract of a preliminary version ofpart of this paper appeared in [22]. Research supported in part by a Grant-in-Aid ofthe Ministry of Education, Science, Sports and Culture ofJapan. Research supported by an NSERC Operating Grant.Part of this research was done during a sabbatical leave atCornell SORIE.§ Research supported in part by a Grant-in-Aid ofthe Ministry of Education, Science, Sports and Culture ofJapan. |