Download:
|
by Jfg De Freitas, M Niranjan, Ah Gee
Journal of VLSI Signal Processing Systems 26(1/2): 119
http://www.cs.berkeley.edu/~jfgf/vlsi2.ps.gz
Add To MetaCart
Abstract:
In this paper, we derive an EM algorithm for nonlinear state space models. We use it to estimate jointly the neural network weights, the model uncertainty and the noise in the data. In the E-step we apply a forward-backward Rauch-Tung-Striebel smoother to compute the network weights. For the M-step, we derive expressions to compute the model uncertainty and the measurement noise. We find that the method is intrinsically very powerful, simple and stable. 1
Citations
|
4704
|
Maximum likelihood from incomplete data via the EM algorithm
– Dempster, Laird, et al.
- 1977
|
|
578
|
Bayesian Data Analysis
– Gelman, Carlin, et al.
- 1995
|
|
444
|
Bayesian Learning for Neural Networks
– Neal
- 1996
|
|
378
|
Stochastic Processes and Filtering Theory
– Jazwinski
- 1970
|
|
323
|
A practical Bayesian framework for backpropagation networks
– MacKay
- 1992
|
|
156
|
A unifying review of linear Gaussian models
– Roweis, Ghahramani
- 1999
|
|
117
|
An approach to time series smoothing and forecasting using the EM algorithm
– Shumway, Stoffer
- 1982
|
|
69
|
Dynamic model of visual recognition predicts neural response properties in the visual cortex
– Rao, Ballard
- 1997
|
|
59
|
Learning dynamic Bayesian networks
– Ghahramani
- 1998
|
|
53
|
ML estimation of a stochastic linear system with the EM algorithm and its application to speech recognition
– Digalakis, Rohlicek, et al.
- 1993
|
|
52
|
Maximum likelihood estimates of linear dynamic systems
– Rauch, Tung, et al.
- 1965
|
|
52
|
Training Multilayer Perceptrons with the Extended Kalman Algorithm
– Singhal, Wu
- 1989
|
|
45
|
Decoupled Extended Kalman Filter Training of Feedforward Layered Networks
– Puskorius, Feldkamp
- 1991
|
|
42
|
Dynamic linear models with switching
– Shumway, Stoffer
- 1991
|
|
37
|
Bayesian radial basis functions of variable dimension
– Holmes, Mallick
- 1998
|
|
25
|
Optimal filtering algorithms for fast learning in feedforward neural networks
– Shah, Palmieri, et al.
- 1992
|
|
17
|
Novelty, confidence and errors in connectionist systems
– Roberts, Penny, et al.
- 1996
|
|
15
|
Feedforward neural networks for nonparametric regression
– Insua, D, et al.
- 1998
|
|
13
|
Robust full Bayesian learning for neural networks
– Andrieu, Freitas, et al.
- 1999
|
|
13
|
Hierarchical Bayesian-Kalman Models for Regularisation and ARD in Sequential Learning
– Freitas, Niranjan, et al.
- 1997
|
|
10
|
Learning dynamical models using expectation-maximisation
– North, Blake
- 1998
|
|
10
|
Bayesian neural networks for classification: How useful is the evidence framework
– Penny, Roberts
- 1998
|
|
10
|
A comparison of fast Fourier transform (FFT) and autoregressive (AR) spectral estimation techniques for the analysis of tremor data
– Spyers-Ashby, Bain, et al.
- 1998
|
|
8
|
Recent extensions to the EM algorithm
– Meng, Rubin
- 1992
|
|
7
|
The EM algorithm to the multiple indicators and multiple causes model via the estimation of the latent variable
– Chen
- 1981
|
