Tensor-variate Restricted Boltzmann Machines

Tu Dinh Nguyen, Truyen Tran, Dinh Phung, Svetha Venkatesh

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

17 Citations (Scopus)

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.

Original languageEnglish
Title of host publicationProceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence
EditorsBlai Bonet, Sven Koenig
Place of PublicationPalo Alto CA USA
PublisherAssociation for the Advancement of Artificial Intelligence (AAAI)
Pages2887-2893
Number of pages7
Volume4
ISBN (Electronic)9781577357025
Publication statusPublished - 2015
Externally publishedYes
EventAAAI Conference on Artificial Intelligence 2015 - Hyatt Regency, Austin, United States of America
Duration: 25 Jan 201530 Jan 2015
Conference number: 29th
http://www.aaai.org/Conferences/AAAI/aaai15.php

Conference

ConferenceAAAI Conference on Artificial Intelligence 2015
Abbreviated titleAAAI 2015
CountryUnited States of America
CityAustin
Period25/01/1530/01/15
OtherCo-located with the 27th Innovative Applications of Artificial Intelligence Conference. Papers at the AAAI 2015 conference will be related here. Any papers presented at the IAAI 2015 part of the conference will be related to that event. The two conferences should have a "relation" to each other put in place to recognise that the conferences were combined into one proceedings.
Internet address

Cite this