Ancestor tree for arbitrary multi-terminal cut functions

AbstractIn many applications, a function is defined on the cuts of a network. In the max-flow min-cut theorem, the function on a cut is simply the sum of all capacities of edges across the cut, and we want the minimum value of a cut separating a given pair of nodes. To find the minimum cuts separating $$(\begin{array}{*{20}c} n \\ 2 \\ \end{array} )$$ pairs of nodes, we only needn − 1 computations to construct the cut-tree. In general, we can define arbitrary values associated with all cuts in a network, and assume that there is a routine which gives the minimum cut separating a pair of nodes. To find the minimum cuts separating $$(\begin{array}{*{20}c} n \\ 2 \\ \end{array} )$$ pairs of nodes, we also only needn − 1 routine calls to construct a binary tree which gives all $$(\begin{array}{*{20}c} n \\ 2 \\ \end{array} )$$ minimum partitions. The binary tree is analogous to the cut-tree of Gomory and Hu.

[1]  Refael Hassin Solution Bases of Multiterminal Cut Problems , 1988, Math. Oper. Res..

[2]  T. C. Hu,et al.  Multi-Terminal Network Flows , 1961 .

[3]  Farhad Shahrokhi,et al.  The maximum concurrent flow problem , 1990, JACM.

[4]  David W. Matula,et al.  Determining edge connectivity in 0(nm) , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[5]  D. R. Fulkerson,et al.  Maximal Flow Through a Network , 1956 .

[6]  Farhad Shahrokhi,et al.  On solving large maximum concurrent flow problems , 1987, CSC '87.

[7]  D. R. Fulkerson,et al.  Flows in Networks. , 1964 .

[8]  Dan Gusfield,et al.  Very Simple Methods for All Pairs Network Flow Analysis , 1990, SIAM J. Comput..