Show simple item record

dc.contributor.authorHendrickx, Julien M.en_US
dc.contributor.authorOlshevsky, Alexen_US
dc.contributor.authorSaligrama, Venkateshen_US
dc.date.accessioned2019-08-30T12:43:26Z
dc.date.available2019-08-30T12:43:26Z
dc.date.issued2019
dc.identifierhttp://arxiv.org/abs/1902.00141v2
dc.identifier.citationJulien M Hendrickx, Alex Olshevsky, Venkatesh Saligrama. "Graph Resistance and Learning from Pairwise Comparisons." Julien Hendrickx, Alexander Olshevsky, Venkatesh Saligrama ; Proceedings of the 36th International Conference on Machine Learning, PMLR 97:2702-2711, 2019.
dc.identifier.urihttps://hdl.handle.net/2144/37589
dc.description.abstractWe consider the problem of learning the qualities of a collection of items by performing noisy comparisons among them. Following the standard paradigm, we assume there is a fixed “comparison graph” and every neighboring pair of items in this graph is compared k times according to the Bradley-Terry-Luce model (where the probability than an item wins a comparison is proportional the item quality). We are interested in how the relative error in quality estimation scales with the comparison graph in the regime where k is large. We show that, asymptotically, the relevant graph-theoretic quantity is the square root of the resistance of the comparison graph. Specifically, we provide an algorithm with relative error decay that scales with the square root of the graph resistance, and provide a lower bound showing that (up to log factors) a better scaling is impossible. The performance guarantee of our algorithm, both in terms of the graph and the skewness of the item quality distribution, significantly outperforms earlier results.en_US
dc.subjectMachine learningen_US
dc.subjectMathematicsen_US
dc.subjectComputer scienceen_US
dc.titleGraph resistance and learning from pairwise comparisonsen_US
dc.typeArticleen_US
dc.description.versionPublished versionen_US
pubs.elements-sourcearxiven_US
pubs.notes15 pages, including 5 pages supplementary materialen_US
pubs.notesEmbargo: Not knownen_US
pubs.organisational-groupBoston Universityen_US
pubs.organisational-groupBoston University, College of Engineeringen_US
pubs.organisational-groupBoston University, College of Engineering, Department of Electrical & Computer Engineeringen_US
dc.identifier.orcid0000-0002-0675-2268 (Saligrama, Venkatesh)
dc.identifier.mycv423994


This item appears in the following Collection(s)

Show simple item record