A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Minimal Learning Machine : Theoretical Results and Clustering-Based Reference Point Selection (2020)


Hämäläinen, Joonas; Alencar, Alisson S. C.; Kärkkäinen, Tommi; Mattos, César L. C.; Souza Júnior, Amauri H.; Gomes, João P. P. (2020). Minimal Learning Machine : Theoretical Results and Clustering-Based Reference Point Selection. Journal of Machine Learning Research, 21, 239. http://jmlr.org/papers/v21/19-786.html


JYU-tekijät tai -toimittajat


Julkaisun tiedot

Julkaisun kaikki tekijät tai toimittajat: Hämäläinen, Joonas; Alencar, Alisson S. C.; Kärkkäinen, Tommi; Mattos, César L. C.; Souza Júnior, Amauri H.; Gomes, João P. P.

Lehti tai sarja: Journal of Machine Learning Research

ISSN: 1532-4435

eISSN: 1533-7928

Julkaisuvuosi: 2020

Volyymi: 21

Artikkelinumero: 239

Kustantaja: JMLR

Julkaisumaa: Yhdysvallat (USA)

Julkaisun kieli: englanti

Pysyvä verkko-osoite: http://jmlr.org/papers/v21/19-786.html

Avoin saatavuus: Open access -julkaisukanavassa ilmestynyt julkaisu

Julkaisu on rinnakkaistallennettu (JYX): https://jyx.jyu.fi/handle/123456789/73804


Tiivistelmä

The Minimal Learning Machine (MLM) is a nonlinear, supervised approach based on learning linear mapping between distance matrices computed in input and output data spaces, where distances are calculated using a subset of points called reference points. Its simple formulation has attracted several recent works on extensions and applications. In this paper, we aim to address some open questions related to the MLM. First, we detail the theoretical aspects that assure the MLM's interpolation and universal approximation capabilities, which had previously only been empirically verified. Second, we identify the major importance of the task of selecting reference points for the MLM's generalization capability. Several clustering-based methods for reference point selection in regression scenarios are then proposed and analyzed. Based on an extensive empirical evaluation, we conclude that the evaluated methods are both scalable and useful. Specifically, for a small number of reference points, the clustering-based methods outperform the standard random selection of the original MLM formulation.


YSO-asiasanat: koneoppiminen; interpolointi; approksimointi

Vapaat asiasanat: Minimal Learning Machine; universal approximation; clustering; reference point selection


Liittyvät organisaatiot


Hankkeet, joissa julkaisu on tehty


OKM-raportointi: Kyllä

Alustava JUFO-taso: 3


Viimeisin päivitys 2021-26-01 klo 12:13