On Bisubmodular Maximization

Bisubmodularity extends the concept of submodularity to set functions with two arguments. We show how bisubmodular maximization leads to richer value-of-information problems, using examples in sensor placement and feature selection. We present the first constant-factor approximation algorithm for a wide class of bisubmodular maximizations.

• Type: Journal
• Year: 2012
• Where: JMLR
• Authors: Ajit Singh, Andrew Guillory, Jeff Bilmes
• Keywords: Approximation Algorithm, Information Problems, JMLR 2012, Programming Languages, Sensor Placement