%0 Conference Proceedings
%T Boltzmann Machine and its Applications in Image Recognition
%+ China University of Mining and Technology (CUMT)
%+ Chinese Academy of Sciences [Changchun Branch] (CAS)
%A Ding, Shifei
%A Zhang, Jian
%A Zhang, Nan
%A Hou, Yanlu
%Z Part 3: Deep Learning
%< avec comité de lecture
%( IFIP Advances in Information and Communication Technology
%B 9th International Conference on Intelligent Information Processing (IIP)
%C Melbourne, VIC, Australia
%3 Intelligent Information Processing VIII
%V AICT-486
%P 108-118
%8 2016-11-18
%D 2016
%R 10.1007/978-3-319-48390-0_12
%K RBM
%K DBM
%K DBN
%K Weight uncertainty
%Z Computer Science [cs]Conference papers
%X The overfitting problems commonly exist in neural networks and RBM models. In order to alleviate the overfitting problem, lots of research has been done. This paper built Weight uncertainty RBM model based on maximum likelihood estimation. And in the experimental section, this paper verified the effectiveness of the Weight uncertainty Deep Belief Network and the Weight uncertainty Deep Boltzmann Machine. In order to improve the images recognition ability, we introduce the spike-and-slab RBM (ssRBM) to our Weight uncertainty RBM and then build the Weight uncertainty spike-and-slab Deep Boltzmann Machine (wssDBM). The experiments showed that, the Weight uncertainty RBM, Weight uncertainty DBN and Weight uncertainty DBM were effective compared with the dropout method. At last, we validate the effectiveness of wssDBM in experimental section.
%G English
%Z TC 12
%2 https://inria.hal.science/hal-01614991/document
%2 https://inria.hal.science/hal-01614991/file/433802_1_En_12_Chapter.pdf
%L hal-01614991
%U https://inria.hal.science/hal-01614991
%~ IFIP
%~ IFIP-AICT
%~ IFIP-TC
%~ IFIP-TC12
%~ IFIP-IIP
%~ IFIP-AICT-486