Computer Science – Discrete Mathematics
Scientific paper
2008-11-12
Discrete Applied Mathematics 157(15) (2009) 3347-3358
Computer Science
Discrete Mathematics
16 pages
Scientific paper
10.1016/j.dam.2009.07.001
It has previously been an open problem whether all Boolean submodular functions can be decomposed into a sum of binary submodular functions over a possibly larger set of variables. This problem has been considered within several different contexts in computer science, including computer vision, artificial intelligence, and pseudo-Boolean optimisation. Using a connection between the expressive power of valued constraints and certain algebraic properties of functions, we answer this question negatively. Our results have several corollaries. First, we characterise precisely which submodular functions of arity 4 can be expressed by binary submodular functions. Next, we identify a novel class of submodular functions of arbitrary arities which can be expressed by binary submodular functions, and therefore minimised efficiently using a so-called expressibility reduction to the Min-Cut problem. More importantly, our results imply limitations on this kind of reduction and establish for the first time that it cannot be used in general to minimise arbitrary submodular functions. Finally, we refute a conjecture of Promislow and Young on the structure of the extreme rays of the cone of Boolean submodular functions.
Cohen David A.
Jeavons Peter G.
Zivny Stanislav
No associations
LandOfFree
The Expressive Power of Binary Submodular Functions does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with The Expressive Power of Binary Submodular Functions, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and The Expressive Power of Binary Submodular Functions will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-101062