Tensor-variate restricted boltzmann machines
dc.contributor.author | Nguyen, T. | |
dc.contributor.author | Tran, The Truyen | |
dc.contributor.author | Phung, D. | |
dc.contributor.author | Venkatesh, S. | |
dc.date.accessioned | 2017-01-30T13:37:37Z | |
dc.date.available | 2017-01-30T13:37:37Z | |
dc.date.created | 2016-03-27T19:30:19Z | |
dc.date.issued | 2015 | |
dc.identifier.citation | Nguyen, T. and Tran, T.T. and Phung, D. and Venkatesh, S. 2015. Tensor-variate restricted boltzmann machines, in Proceedings of the National Conference on Artificial Intelligence, pp. 2887-2893. | |
dc.identifier.uri | http://hdl.handle.net/20.500.11937/33530 | |
dc.description.abstract |
Restricted Boltzmann Machines (RBMs) are an important class of latent variable models for representing vector data. An under-explored area is multimode data, where each data point is a matrix or a tensor. Standard RBMs applying to such data would require vectorizing matrices and tensors, thus resulting in unnecessarily high dimensionality and at the same time, destroying the inherent higher-order interaction structures. This paper introduces Tensor-variate Restricted Boltzmann Machines (TvRBMs) which generalize RBMs to capture the multiplicative interaction between data modes and the latent variables. TvRBMs are highly compact in that the number of free parameters grows only linear with the number of modes. We demonstrate the capacity of TvRBMs on three real-world applications: handwritten digit classification, face recognition and EEG-based alcoholic diagnosis. The learnt features of the model are more discriminative than the rivals, resulting in better classification performance. | |
dc.title | Tensor-variate restricted boltzmann machines | |
dc.type | Conference Paper | |
dcterms.source.volume | 4 | |
dcterms.source.startPage | 2887 | |
dcterms.source.endPage | 2893 | |
dcterms.source.title | Proceedings of the National Conference on Artificial Intelligence | |
dcterms.source.series | Proceedings of the National Conference on Artificial Intelligence | |
dcterms.source.isbn | 9781577357025 | |
curtin.department | Multi-Sensor Proc & Content Analysis Institute | |
curtin.accessStatus | Fulltext not available |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |