Skip to main navigation Skip to search Skip to main content

Matching Map Recovery with an Unknown Number of Outliers

  • Russian-Armenian University
  • Yerevan State University

Research output: Contribution to journalConference articlepeer-review

Abstract

We consider the problem of finding the matching map between two sets of d-dimensional noisy feature-vectors. The distinctive feature of our setting is that we do not assume that all the vectors of the first set have their corresponding vector in the second set. If n and m are the sizes of these two sets, we assume that the matching map that should be recovered is defined on a subset of unknown cardinality k ≤ min(n, m). We show that, in the high-dimensional setting, if the signal-to-noise ratio is larger than 5(dlog(4nm/α))1/4, then the true matching map can be recovered with probability 1 - α. Interestingly, this threshold does not depend on k and is the same as the one obtained in prior work in the case of k = min(n, m). The procedure for which the aforementioned property is proved is obtained by a data-driven selection among candidate mappings {π(equation presented)k : k ∈ [min(n, m)]}. Each π(equation presented)k minimizes the sum of squares of distances between two sets of size k. The resulting optimization problem can be formulated as a minimum-cost flow problem, and thus solved efficiently. Finally, we report the results of numerical experiments on both synthetic and real-world data that illustrate our theoretical results and provide further insight into the properties of the algorithms studied in this work.

Original languageEnglish
Pages (from-to)891-906
Number of pages16
JournalProceedings of Machine Learning Research
Volume206
Publication statusPublished - 1 Jan 2023
Event26th International Conference on Artificial Intelligence and Statistics, AISTATS 2023 - Valencia, Spain
Duration: 25 Apr 202327 Apr 2023

Fingerprint

Dive into the research topics of 'Matching Map Recovery with an Unknown Number of Outliers'. Together they form a unique fingerprint.

Cite this