Submodular Hamming Metrics

Abstract

We show that there is a largely unexplored class of functions (positive polymatroids) that can define proper discrete metrics over pairs of binary vectors and that are fairly tractable to optimize over. By exploiting submodularity, we are able to give hardness results and approximation algorithms for optimizing over such metrics. Additionally, we demonstrate empirically the effectiveness of these metrics and associated algorithms on both a metric minimization task (a form of clustering) and also a metric maximization task (generating diverse k-best lists).

15 pages, 1 figure, a short version of this will appear in the NIPS 2015 conference

References in corpus (1)

• Distributed Representations of Words and Phrases and their Compositionality

Cited by in corpus (6)

• Submodular Combinatorial Information Measures with Applications in Machine Learning
• A note on the triangle inequality for the Jaccard distance
• Clustering with a Reject Option: Interactive Clustering as Bayesian Prior Elicitation
• A Convex Surrogate Operator for General Non-Modular Loss Functions
• Finding Submodularity Hidden in Symmetric Difference
• Supermodular Locality Sensitive Hashes