Conference proceeding
On Deep Multi-View Representation Learning
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 37, Vol.37, pp.1083-1092
Proceedings of Machine Learning Research
01/01/2015
Abstract
We consider learning representations (features) in the setting in which we have access to multiple unlabeled views of the data for representation learning while only one view is available at test time. Previous work on this problem has proposed several techniques based on deep neural networks, typically involving either autoencoder-like networks with a reconstruction objective or paired feedforward networks with a correlation-based objective. We analyze several techniques based on prior work, as well as new variants, and compare them experimentally on visual, speech, and language domains. To our knowledge this is the first head-to-head comparison of a variety of such techniques on multiple tasks. We find an advantage for correlation-based representation learning, while the best results on most tasks are obtained with our new variant, deep canonically correlated autoencoders (DCCAE).
Details
- Title: Subtitle
- On Deep Multi-View Representation Learning
- Creators
- Weiran Wang - Toyota Technol Inst, Chicago, IL 60637 USARaman Arora - Johns Hopkins UniversityKaren Livescu - Toyota Technol Inst, Chicago, IL 60637 USAJeff Bilmes - University of Washington
- Contributors
- F Bach (Editor)D Blei (Editor)
- Resource Type
- Conference proceeding
- Publication Details
- INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 37, Vol.37, pp.1083-1092
- Publisher
- JMLR-JOURNAL MACHINE LEARNING RESEARCH
- Series
- Proceedings of Machine Learning Research
- ISSN
- 2640-3498
- Number of pages
- 10
- Grant note
- IIS-1321015 / NSF; National Science Foundation (NSF)
- Language
- English
- Date published
- 01/01/2015
- Academic Unit
- Computer Science
- Record Identifier
- 9984696578402771
Metrics
1 Record Views