Current browse context:
cs.LG
Change to browse by:
References & Citations
Computer Science > Machine Learning
Title: Submodular Combinatorial Information Measures with Applications in Machine Learning
(Submitted on 27 Jun 2020 (v1), last revised 2 Mar 2021 (this version, v6))
Abstract: Information-theoretic quantities like entropy and mutual information have found numerous uses in machine learning. It is well known that there is a strong connection between these entropic quantities and submodularity since entropy over a set of random variables is submodular. In this paper, we study combinatorial information measures that generalize independence, (conditional) entropy, (conditional) mutual information, and total correlation defined over sets of (not necessarily random) variables. These measures strictly generalize the corresponding entropic measures since they are all parameterized via submodular functions that themselves strictly generalize entropy. Critically, we show that, unlike entropic mutual information in general, the submodular mutual information is actually submodular in one argument, holding the other fixed, for a large class of submodular functions whose third-order partial derivatives satisfy a non-negativity property. This turns out to include a number of practically useful cases such as the facility location and set-cover functions. We study specific instantiations of the submodular information measures on these, as well as the probabilistic coverage, graph-cut, and saturated coverage functions, and see that they all have mathematically intuitive and practically useful expressions. Regarding applications, we connect the maximization of submodular (conditional) mutual information to problems such as mutual-information-based, query-based, and privacy-preserving summarization -- and we connect optimizing the multi-set submodular mutual information to clustering and robust partitioning.
Submission history
From: Rishabh Iyer [view email][v1] Sat, 27 Jun 2020 17:45:32 GMT (229kb,D)
[v2] Sat, 4 Jul 2020 03:43:56 GMT (229kb,D)
[v3] Tue, 22 Dec 2020 22:26:01 GMT (221kb,D)
[v4] Tue, 29 Dec 2020 07:14:34 GMT (215kb)
[v5] Fri, 1 Jan 2021 02:52:27 GMT (14741kb)
[v6] Tue, 2 Mar 2021 19:58:48 GMT (54kb)
Link back to: arXiv, form interface, contact.