A4 Artikkeli konferenssijulkaisussa
Group Nonnegative Matrix Factorization with Sparse Regularization in Multi-set Data (2020)


Wang, Xiulin; Liu, Wenya; Cong, Fengyu; Ristaniemi, Tapani (2020). Group Nonnegative Matrix Factorization with Sparse Regularization in Multi-set Data. In EUSIPCO 2020 : 28th European Signal Processing Conference (pp. 2125-2129). European Signal Processing Conference. IEEE. DOI: 10.23919/Eusipco47968.2020.9287756


JYU-tekijät tai -toimittajat


Julkaisun tiedot

Julkaisun kaikki tekijät tai toimittajat: Wang, Xiulin; Liu, Wenya; Cong, Fengyu; Ristaniemi, Tapani

Emojulkaisu: EUSIPCO 2020 : 28th European Signal Processing Conference

Konferenssin paikka ja aika: Amsterdam, Netherlands, 18.-21.1.2021

ISBN: 978-1-7281-5001-7

eISBN: 978-9-0827-9705-3

Lehti tai sarja: European Signal Processing Conference

ISSN: 2219-5491

eISSN: 2076-1465

Julkaisuvuosi: 2020

Artikkelin sivunumerot: 2125-2129

Kustantaja: IEEE

Julkaisumaa: Yhdysvallat (USA)

Julkaisun kieli: englanti

DOI: https://doi.org/10.23919/Eusipco47968.2020.9287756

Avoin saatavuus: Julkaisukanava ei ole avoin

Julkaisukanavan avoin saatavuus:

Julkaisun avoin saatavuus:


Tiivistelmä

Constrained joint analysis of data from multiplesources has received widespread attention for that it allowsus to explore potential connections and extract meaningful hidden components. In this paper, we formulate a flexible jointsource separation model termed as group nonnegative matrix factorization with sparse regularization (GNMF-SR), which aimsto jointly analyze the partially coupled multi-set data. In the GNMF-SR model, common and individual patterns of particular underlying factors can be extracted simultaneously with imposing nonnegative constraint and sparse penalty. Alternating optimization and alternating direction method of multipliers (ADMM) are combined to solve the GNMF-SR model. Using the experimentof simulated fMRI-like data, we demonstrate the ADMM-basedGNMF-SR algorithm can achieve the better performance.


YSO-asiasanat: signaalinkäsittely; koneoppiminen; algoritmit

Vapaat asiasanat: Alternating direction method of multipliers; coupled; group nonnegative matrix factorization; joint analysis; sparse representation


Liittyvät organisaatiot


OKM-raportointi: Kyllä

Alustava JUFO-taso: 1


Viimeisin päivitys 2020-28-12 klo 16:05