Latent Space Factorisation and Manipulation via Matrix Subspace Projection
- Submitting institution
-
The University of Surrey
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 9027031_2
- Type
- E - Conference contribution
- DOI
-
-
- Title of conference / published proceedings
- 37th International Conference on Machine Learning
- First page
- 5916
- Volume
- 119
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- -
- Year of publication
- 2020
- URL
-
http://proceedings.mlr.press/v119/
- 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
-
-
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The approach is the first to get good results when trained to disentangle many attributes simultaneously. It is also universal to apply to different autoencoders. All code and data are available and experiments reproducible. The work has been taken up by colleagues in Shanghai Jiao Tong University (dingyue@sjtu.edu.cn) for work on building models for object pose tracking.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -