Tensor decomposition via joint matrix schur decomposition
- Submitting institution
-
Royal Holloway and Bedford New College
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 34725081
- Type
- E - Conference contribution
- DOI
-
10.5555/3045390.3045687
- Title of conference / published proceedings
- ICML'16 : Proceedings of the 33rd International Conference on International Conference on Machine Learning
- First page
- 2820
- Volume
- 48
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- June
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper presents a novel approach to tensor decomposition based on the simultaneous Schur decomposition of several matrices and includes theoretical and empirical results. Tensor decomposition is a popular ML topic and the research community has appreciated the novelty and originality of the proposed method. The paper was presented at a major ML conferences (ICML 2016, acceptance rate 24%) and is now cited together with very well-known works on the subject. The proposed approach has inspired further analysis of related more general problems, e.g. see Colombo et al. "A posteriori error bounds for joint matrix decomposition problems'' (NIPS 2016).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -