Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published October 1996 | Published
Journal Article Open

Hybrid modeling, HMM/NN architectures, and protein applications

Abstract

We describe a hybrid modeling approach where the parameters of a model are calculated and modulated by another model, typically a neural network (NN), to avoid both overfitting and underfitting. We develop the approach for the case of Hidden Markov Models (HMMs), by deriving a class of hybrid HMM/NN architectures. These architectures can be trained with unified algorithms that blend HMM dynamic programming with NN backpropagation. In the case of complex data, mixtures of HMMs or modulated HMMs must be used. NNs can then be applied both to the parameters of each single HMM, and to the switching or modulation of the models, as a function of input or context. Hybrid HMM/NN architectures provide a flexible NN parameterization for the control of model structure and complexity. At the same time, they can capture distributions that, in practice, are inaccessible to single HMMs. The HMM/NN hybrid approach is tested, in its simplest form, by constructing a model of the immunoglobulin protein family. A hybrid model is trained, and a multiple alignment derived, with less than a fourth of the number of parameters used with previous single HMMs.

Additional Information

© 1996 The MIT Press. Received February 2, 1995, accepted February 8, 1996. Posted Online April 4, 2008. The work of P.B. is supported by a grant from the ONR. The work of Y.C. is supported in part by Grant R43 LM05780 from the National Library of Medicine. The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the National Library of Medicine.

Attached Files

Published - BALnc96.pdf

Files

BALnc96.pdf
Files (1.3 MB)
Name Size Download all
md5:afd1723b83e34ba0df5b2d33f303720d
1.3 MB Preview Download

Additional details

Created:
August 22, 2023
Modified:
October 17, 2023