Open access
Datum
2019-06-01Typ
- Conference Paper
Abstract
We show, given a binary integer function diamond that is piecewise polynomial, that (+,diamond) vector products are equivalent under one-to-polylog reductions to the computation of the Hamming distance. Examples include the dominance and l_{2p+1} distances for constant p. Our results imply equivalence (up to polylog factors) between the complexity of computing All Pairs Hamming Distance, All Pairs l_{2p+1} Distance and Dominance Matrix Product, and equivalence between Hamming Distance Pattern Matching, l_{2p+1} Pattern Matching and Less-Than Pattern Matching. The resulting algorithms for l_{2p+1} Pattern Matching and All Pairs l_{2p+1}, for 2p+1 = 3,5,7,... are likely to be optimal, given lack of progress in improving upper bounds for Hamming distance in the past 30 years. While reductions between selected pairs of products were presented in the past, our work is the first to generalize them to a general class of functions, showing that a wide class of "intermediate" complexity problems are in fact equivalent. Mehr anzeigen
Persistenter Link
https://doi.org/10.3929/ethz-b-000351621Publikationsstatus
publishedExterne Links
Buchtitel
30th Annual Symposium on Combinatorial Pattern Matching (CPM 2019)Zeitschrift / Serie
Leibniz International Proceedings in Informatics (LIPIcs)Band
Seiten / Artikelnummer
Verlag
Schloss Dagstuhl - Leibniz-Zentrum für InformatikKonferenz
Thema
Fine grained complexity; Approximate pattern matching; Matrix products