This is more proof of concept then actual practical implementation. Its runtime would improve the current state of art. See http://www.mit.edu/~aram/talks/10-in

This is more proof of concept then actual practical implementation. Its runtime would improve the current state of art. See http://www.mit.edu/~aram/talks/10-invert-rutgers-print.pdf http://www.mit.edu/~aram/talks/10-invert-rutgers-print.pdf