Issue No. 08 - Aug. (2013 vol. 35)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TPAMI.2013.19
Bo Chen , Duke University, Durham
Gungor Polatkan , Princeton University, Princeton
Guillermo Sapiro , Duke University, Durham
David Blei , Princeton University, Princeton
David Dunson , Duke University, Durham
Lawrence Carin , Duke University, Durham
Unsupervised multilayered (“deep”) models are considered for imagery. The model is represented using a hierarchical convolutional factor-analysis construction, with sparse factor loadings and scores. The computation of layer-dependent model parameters is implemented within a Bayesian setting, employing a Gibbs sampler and variational Bayesian (VB) analysis that explicitly exploit the convolutional nature of the expansion. To address large-scale and streaming data, an online version of VB is also developed. The number of dictionary elements at each layer is inferred from the data, based on a beta-Bernoulli implementation of the Indian buffet process. Example results are presented for several image-processing applications, with comparisons to related models in the literature.
Dictionaries, Convolution, Computational modeling, Mathematical model, Analytical models, Load modeling, Bayesian methods, factor analysis, Bayesian, deep learning, convolutional, dictionary learning
L. Carin, D. Dunson, D. Blei, G. Sapiro, B. Chen and G. Polatkan, "Deep Learning with Hierarchical Convolutional Factor Analysis," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 35, no. , pp. 1887-1901, 2013.