Comparative Study of Inference Methods for Bayesian Nonnegative Matrix Factorisation
| cam.issuedOnline | 2017-12-30 | |
| dc.contributor.author | Brouwer, T | |
| dc.contributor.author | Frellsen, J | |
| dc.contributor.author | Lió, P | |
| dc.contributor.orcid | Lio, Pietro [0000-0002-0540-5053] | |
| dc.date.accessioned | 2018-10-25T05:02:03Z | |
| dc.date.available | 2018-10-25T05:02:03Z | |
| dc.date.issued | 2017 | |
| dc.description.abstract | In this paper, we study the trade-offs of different inference approaches for Bayesian matrix factorisation methods, which are commonly used for predicting missing values, and for finding patterns in the data. In particular, we consider Bayesian nonnegative variants of matrix factorisation and tri-factorisation, and compare non-probabilistic inference, Gibbs sampling, variational Bayesian inference, and a maximum-a-posteriori approach. The variational approach is new for the Bayesian nonnegative models. We compare their convergence, and robustness to noise and sparsity of the data, on both synthetic and real-world datasets. Furthermore, we extend the models with the Bayesian automatic relevance determination prior, allowing the models to perform automatic model selection, and demonstrate its efficiency. | |
| dc.identifier.doi | 10.17863/CAM.31752 | |
| dc.identifier.eissn | 1611-3349 | |
| dc.identifier.issn | 0302-9743 | |
| dc.identifier.uri | https://www.repository.cam.ac.uk/handle/1810/284376 | |
| dc.language.iso | eng | |
| dc.publisher | Springer International Publishing | |
| dc.publisher.url | http://dx.doi.org/10.1007/978-3-319-71249-9_31 | |
| dc.subject | stat.ML | |
| dc.subject | stat.ML | |
| dc.subject | cs.LG | |
| dc.title | Comparative Study of Inference Methods for Bayesian Nonnegative Matrix Factorisation | |
| dc.type | Article | |
| prism.endingPage | 529 | |
| prism.publicationDate | 2017 | |
| prism.publicationName | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | |
| prism.startingPage | 513 | |
| prism.volume | 10534 LNAI | |
| pubs.funder-project-id | Engineering and Physical Sciences Research Council (EP/M506485/1) | |
| rioxxterms.licenseref.startdate | 2017-01-01 | |
| rioxxterms.licenseref.uri | http://www.rioxx.net/licenses/all-rights-reserved | |
| rioxxterms.type | Journal Article/Review | |
| rioxxterms.version | AM | |
| rioxxterms.versionofrecord | 10.1007/978-3-319-71249-9_31 |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- 1707.05147v1.pdf
- Size:
- 5.34 MB
- Format:
- Adobe Portable Document Format
- Description:
- Accepted version
- Licence
- http://www.rioxx.net/licenses/all-rights-reserved
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- DepositLicenceAgreement.pdf
- Size:
- 417.78 KB
- Format:
- Adobe Portable Document Format