Szczegóły publikacji
Opis bibliograficzny
GPU-embedding of kNN-graph representing large and high-dimensional data / Bartosz MINCH, Mateusz Nowak, Rafał WCISŁO, Witold DZWINEL // W: Computational Science - ICCS 2020 : 20th International Conference : Amsterdam, The Netherlands, June 3–5, 2020 : proceedings, Pt. 2 / eds. Valeria V. Krzhizhanovskaya, [et al.]. — Cham : Springer Nature Switzerland, cop. 2020. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; LNCS 12138. Theoretical Computer Science and General Issues ; ISSN 0302-9743). — ISBN: 978-3-030-50416-8; e-ISBN: 978-3-030-50417-5. — S. 322–336. — Bibliogr. s. 335–336, Abstr. — Publikacja dostępna online od: 2020-06-15
Autorzy (4)
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 129150 |
---|---|
Data dodania do BaDAP | 2020-06-25 |
Tekst źródłowy | URL |
DOI | 10.1007/978-3-030-50417-5_24 |
Rok publikacji | 2020 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Wydawca | Springer |
Konferencja | 20th International Conference on Computational Science |
Czasopisma/serie | Theoretical Computer Science and General Issues, Lecture Notes in Computer Science |
Abstract
Interactive visual exploration of large and multidimensional data still needs more efficient ND→2D data embedding (DE) algorithms. We claim that the visualization of very high-dimensional data is equivalent to the problem of 2D embedding of undirected kNN-graphs. We demonstrate that high quality embeddings can be produced with minimal time&memory complexity. A very efficient GPU version of IVHD (interactive visualization of high-dimensional data) algorithm is presented, and we compare it to the state-of-the-art GPU-implemented DE methods: BH-SNE-CUDA and AtSNE-CUDA. We show that memory and time requirements for IVHD-CUDA are radically lower than those for the baseline codes. For example, IVHD-CUDA is almost 30 times faster in embedding (without the procedure of kNN graph generation, which is the same for all the methods) of the largest ( M=1.4⋅106 ) YAHOO dataset than AtSNE-CUDA. We conclude that in the expense of minor deterioration of embedding quality, compared to the baseline algorithms, IVHD well preserves the main structural properties of ND data in 2D for radically lower computational budget. Thus, our method can be a good candidate for a truly big data ( M=108+ ) interactive visualization.