Dual Simplex Volume Maximization for Simplex-Structured Matrix Factorization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916557427310592 |
|---|---|
| author | Abdolali, Maryam Barbarino, Giovanni Gillis, Nicolas |
| author_facet | Abdolali, Maryam Barbarino, Giovanni Gillis, Nicolas |
| contents | Simplex-structured matrix factorization (SSMF) is a generalization of nonnegative matrix factorization, a fundamental interpretable data analysis model, and has applications in hyperspectral unmixing and topic modeling. To obtain identifiable solutions, a standard approach is to find minimum-volume solutions. By taking advantage of the duality/polarity concept for polytopes, we convert minimum-volume SSMF in the primal space to a maximum-volume problem in the dual space. We first prove the identifiability of this maximum-volume dual problem. Then, we use this dual formulation to provide a novel optimization approach which bridges the gap between two existing families of algorithms for SSMF, namely volume minimization and facet identification. Numerical experiments show that the proposed approach performs favorably compared to the state-of-the-art SSMF algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_20197 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Dual Simplex Volume Maximization for Simplex-Structured Matrix Factorization Abdolali, Maryam Barbarino, Giovanni Gillis, Nicolas Numerical Analysis Information Retrieval Machine Learning Signal Processing Simplex-structured matrix factorization (SSMF) is a generalization of nonnegative matrix factorization, a fundamental interpretable data analysis model, and has applications in hyperspectral unmixing and topic modeling. To obtain identifiable solutions, a standard approach is to find minimum-volume solutions. By taking advantage of the duality/polarity concept for polytopes, we convert minimum-volume SSMF in the primal space to a maximum-volume problem in the dual space. We first prove the identifiability of this maximum-volume dual problem. Then, we use this dual formulation to provide a novel optimization approach which bridges the gap between two existing families of algorithms for SSMF, namely volume minimization and facet identification. Numerical experiments show that the proposed approach performs favorably compared to the state-of-the-art SSMF algorithms. |
| title | Dual Simplex Volume Maximization for Simplex-Structured Matrix Factorization |
| topic | Numerical Analysis Information Retrieval Machine Learning Signal Processing |
| url | https://arxiv.org/abs/2403.20197 |