Abstract
With the development of deep neural networks, the model of restricted Boltzmann machine(RBM) has gradually become one of the essential aspects in deep learning researches. Because of the presence of the partition function, it is intractable to get the model selection, control the complexity, and learn an exact maximum likelihood in RBM model. A kind of effective measure is approximate inference that adopts annealing importance sampling(AIS) scheme only to evaluate a RBM’s performance. At present, there is little quantitative analysis on discrepancies generated by different RBM models. So we focus on the innovation research on some quantitative evaluation of the generalization performance of all kinds of sparse RBM models, including the classical sparse RBM(SpRBM) and the log sum sparse RBM(LogSumRBM). We discuss the influence and efficiency of the AIS strategy for these sparse RBMs’ estimations. Particularly, we confirm that the LogSumRBM is the optimal model in RBM and sparse RBMs for its smaller deviations in the assessment results regardless of the training MNIST data and the test, which provides a guarantee on some theories and experience in the choice of the deep learning models in the future.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Asja, F., Christian, I.: Bounding the bias of contrastive diverence learning. Neural Comput. 23(3), 664–673 (2011)
Athanasios, K.N., Ben, J.A.K.: Deep belief networks for dimensionality reduction, vol. 20, pp. 185–191 (2008)
Geoffrey, E.H., Ruslan, S.: Reducing the dimensionality of data with nerual networks. Science 313, 504–507 (2006). American Association for the Advancement of Science
Geoffrey, E.H., Simon, O., Yee-Whye, T.: A fast learning algorithm for deep belief nets. Neural Comput. 18, 1527–1554 (2006). Massachusetts Institute of Technology Press
Honglak, L., Chaitanya, E., Andrew, Y.N.: Sparse deep belief net model for visual area v2. In: Advances in Neural Information Processing Systems, vol. 20, pp. 873–880. DBLP (2007)
Jakub, M.T., Adam, G.: Sparse hidden units activation in restricted boltzmann machine, pp. 181–185 (2015)
Ji, N., Zhang, J., Zhang, C., Yin, Q.: Enhancing performance of restricted boltzmann machines via log-sum regularization. Knowl. Based Syst. 63(3), 82–96 (2014)
Ruslan, S., Iain, M.: On the quantitative analysis of deep belief networks, pp. 872–879 (2008)
Vinod, N., Geoffrey, E.H.: 3d object recognition with deep belief nets. In: Advances in Neural Information Processing Systems, vol. 22, pp. 1527–1554. DBLP (2012)
Yoshua, B., Olivier, D.: Justifying and generalizing contrastive divergence. Neural Comput. 21(6), 1601–1621 (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing AG
About this paper
Cite this paper
Zhang, Y. et al. (2016). On the Quantitative Analysis of Sparse RBMs. In: Chen, E., Gong, Y., Tie, Y. (eds) Advances in Multimedia Information Processing - PCM 2016. PCM 2016. Lecture Notes in Computer Science(), vol 9916. Springer, Cham. https://doi.org/10.1007/978-3-319-48890-5_44
Download citation
DOI: https://doi.org/10.1007/978-3-319-48890-5_44
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-48889-9
Online ISBN: 978-3-319-48890-5
eBook Packages: Computer ScienceComputer Science (R0)