On the Approximation Relationship between Optimizing Ratio of Submodular (RS) and Difference of Submodular (DS) Functions

Abstract

We demonstrate that from an algorithm guaranteeing an approximation factor for the ratio of submodular (RS) optimization problem, we can build another algorithm having a different kind of approximation guarantee -- weaker than the classical one -- for the difference of submodular (DS) optimization problem, and vice versa. We also illustrate the link between these two problems by analyzing a \textsc{Greedy} algorithm which approximately maximizes objective functions of the form $\Psi (f,g)$, where $f,g$ are two non-negative, monotone, submodular functions and $\Psi$ is a {quasiconvex} 2-variables function, which is non decreasing with respect to the first variable. For the choice $\Psi (f,g)\triangleq f/g$, we recover RS, and for the choice $\Psi (f,g)\triangleq f-g$, we recover DS. To the best of our knowledge, this greedy approach is new for DS optimization. For RS optimization, it reduces to the standard \textsc{GreedRatio} algorithm that has already been analyzed previously. However, our analysis is novel for this case.

References in corpus (6)

• Near-optimal Nonmyopic Value of Information in Graphical Models
• Algorithms for Approximate Minimization of the Difference Between Submodular Functions, with Applications
• Submodular Maximization Beyond Non-negativity: Guarantees, Fast Algorithms, and Applications
• Curvature and Optimal Algorithms for Learning and Minimizing Submodular Functions
• A submodular-supermodular procedure with applications to discriminative structure learning
• Exploiting Structure of Uncertainty for Efficient Matroid Semi-Bandits