A new, simple and fast algorithm finds a sequence of nested minimum cuts of a bipartite parametric flow network. Instead of working with the original parametric flow-network, the new method works with a derived non-parametric flow network and finds a particular state of the flows in the derived network. In a single scan of the flows in this special state, the sequence of nested minimum cuts of the original parametric flow network can be written out. The models this new method can solve cover a number of interesting real-world problems.
Download the slides for this presentation in PDF format.
About the speaker:
Bin Zhang has a BS and MS in Mathematics from Nankai University and a Ph.D in Computer Science from Harvard University. He has close to 20 years experience in industrial research and development. After joining HP Labs, he designed computer algorithms in the areas of data mining and information management.