Massively Parallel Correlation Clustering in Bounded Arboricity Graphs

Melanie Cambus, Davin Choo, Havu Miikonen, Jara Uitto

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

7 Sitaatiot (Scopus)
19 Lataukset (Pure)

Abstrakti

Identifying clusters of similar elements in a set is a common task in data analysis. With the immense growth of data and physical limitations on single processor speed, it is necessary to find efficient parallel algorithms for clustering tasks. In this paper, we study the problem of correlation clustering in bounded arboricity graphs with respect to the Massively Parallel Computation (MPC) model. More specifically, we are given a complete graph where the edges are either positive or negative, indicating whether pairs of vertices are similar or dissimilar. The task is to partition the vertices into clusters with as few disagreements as possible. That is, we want to minimize the number of positive inter-cluster edges and negative intra-cluster edges.
Consider an input graph G on n vertices such that the positive edges induce a λ-arboric graph. Our main result is a 3-approximation (in expectation) algorithm to correlation clustering that runs in 𝒪(log λ ⋅ poly(log log n)) MPC rounds in the strongly sublinear memory regime. This is obtained by combining structural properties of correlation clustering on bounded arboricity graphs with the insights of Fischer and Noever (SODA '18) on randomized greedy MIS and the PIVOT algorithm of Ailon, Charikar, and Newman (STOC '05). Combined with known graph matching algorithms, our structural property also implies an exact algorithm and algorithms with worst case (1+ε)-approximation guarantees in the special case of forests, where λ = 1.
AlkuperäiskieliEnglanti
Otsikko35th International Symposium on Distributed Computing, DISC 2021
ToimittajatSeth Gilbert
KustantajaSchloss Dagstuhl - Leibniz-Zentrum für Informatik
Sivumäärä18
ISBN (elektroninen)978-3-95977-210-5
DOI - pysyväislinkit
TilaJulkaistu - 2021
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational Symposium on Distributed Computing - Virtual, Online, Freiburg, Saksa
Kesto: 4 lokak. 20218 lokak. 2021
Konferenssinumero: 35

Julkaisusarja

NimiLeibniz International Proceedings in Informatics, LIPIcs
KustantajaSchloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
Vuosikerta209
ISSN (elektroninen)1868-8969

Conference

ConferenceInternational Symposium on Distributed Computing
LyhennettäDISC
Maa/AlueSaksa
KaupunkiFreiburg
Ajanjakso04/10/202108/10/2021

Sormenjälki

Sukella tutkimusaiheisiin 'Massively Parallel Correlation Clustering in Bounded Arboricity Graphs'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä