Joint M-Best-Diverse Labelings as a Parametric Submodular Minimization

TL;DR

This paper introduces a novel approach to efficiently compute the M-best diverse labelings in graphical models by leveraging parametric submodular minimization, leading to faster and more accurate results.

Contribution

It establishes a theoretical link between diversity in submodular energies and parametric submodular minimization, enabling new efficient algorithms for M-best diverse labelings.

Findings

New algorithms outperform existing methods in speed and accuracy.

Exact M-best diverse labelings are computed faster than approximate methods.

The approach applies to various diversity measures with closed-form parameter computation.

Abstract

We consider the problem of jointly inferring the M-best diverse labelings for a binary (high-order) submodular energy of a graphical model. Recently, it was shown that this problem can be solved to a global optimum, for many practically interesting diversity measures. It was noted that the labelings are, so-called, nested. This nestedness property also holds for labelings of a class of parametric submodular minimization problems, where different values of the global parameter $\gamma$ give rise to different solutions. The popular example of the parametric submodular minimization is the monotonic parametric max-flow problem, which is also widely used for computing multiple labelings. As the main contribution of this work we establish a close relationship between diversity with submodular energies and the parametric submodular minimization. In particular, the joint M-best diverse labelings can be obtained by running a non-parametric submodular minimization (in the special case - max-flow) solver for M different values of $\gamma$ in parallel, for certain diversity measures. Importantly, the values for $\gamma$ can be computed in a closed form in advance, prior to any optimization. These theoretical results suggest two simple yet efficient algorithms for the joint M-best diverse problem, which outperform competitors in terms of runtime and quality of results. In particular, as we show in the paper, the new methods compute the exact M-best diverse labelings faster than a popular method of Batra et al., which in some sense only obtains approximate solutions.