Ordinal embeddings of minimum relaxation: General properties, trees, and ultrametrics

Noga Alon, Mihai Bǎdoiu, Erik D. Demaine, Martin Farach-Colton, Mohammadtaghi Hajiaghayi, Anastasios Sidiropoulos

    Research output: Contribution to conferencePaperpeer-review

    Abstract

    We introduce a new notion of embedding, called minimum-relaxation ordinal embedding, parallel to the standard notion of minimum-distortion (metric) embedding. In an ordinal embedding, it is the relative order between pairs of distances, and not the distances themselves, that must be preserved as much as possible. The (multiplicative) relaxation of an ordinal embedding is the maximum ratio between two distances whose relative order is inverted by the embedding. We develop several worst-case bounds and approximation algorithms on ordinal embedding. In particular, we establish that ordinal embedding has many qualitative differences from metric embedding, and capture the ordinal behavior of ultrametrics and shortest-path metrics of unweighted trees.

    Original languageEnglish (US)
    Pages650-659
    Number of pages10
    StatePublished - 2005
    EventSixteenth Annual ACM-SIAM Symposium on Discrete Algorithms - Vancouver, BC, United States
    Duration: Jan 23 2005Jan 25 2005

    Other

    OtherSixteenth Annual ACM-SIAM Symposium on Discrete Algorithms
    Country/TerritoryUnited States
    CityVancouver, BC
    Period1/23/051/25/05

    ASJC Scopus subject areas

    • Software
    • General Mathematics

    Fingerprint

    Dive into the research topics of 'Ordinal embeddings of minimum relaxation: General properties, trees, and ultrametrics'. Together they form a unique fingerprint.

    Cite this