Sketches are compact bit strings that are considered as products of an LSH for high-dimensional data. We use them in filtering for narrowing down solution candidates in similarity search. We propose a pivot selection method for narrow sketches with a length such as 16-bits by optimization algorithms with the accuracy of filtering itself as the objective function.
CITATION STYLE
Higuchi, N., Imamura, Y., Mic, V., Shinohara, T., Hirata, K., & Kuboyama, T. (2020). Pivot selection for narrow sketches by optimization algorithms. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12440 LNCS, pp. 33–46). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-60936-8_3
Mendeley helps you to discover research relevant for your work.