Multi-agent Diverse Generative Adversarial Networks
- Submitting institution
-
The University of Bath
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 203656490
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR.2018.00888
- Title of conference / published proceedings
- Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
- First page
- 8513
- Volume
- -
- Issue
- -
- ISSN
- 1063-6919
- Open access status
- Deposit exception
- Month of publication
- December
- Year of publication
- 2018
- 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
-
4
- Research group(s)
-
-
- Citation count
- 27
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Published in the top conference in Computer Vision, this work provided a generalisation to Generative Adversarial Networks that enabled it to generate diverse samples, a limitation that was common to most GANs. The discriminator had to identify which of the generators had generated a particular sample. This was also shown theoretically to be optimal. This is now commonly accepted as a standard technique for enabling diversity in GANs and was presented for instance by Philip Isola (MIT) in the tutorial on GANs in CVPR 2018 as one of the standard ways to enhance diversity.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
