Online Variational Inference for State-Space Models with Point-Process Observations

We present a variational Bayesian (VB) approach for the state and parameter inference of a state-space model with point-process observations, a physiologically plausible model for signal processing of spike data. We also give the derivation of a variational smoother, as well as an efficient online filtering algorithm, which can also be used to track changes in physiological parameters. The methods are assessed on simulated data, and results are compared to expectation-maximization, as well as Monte Carlo estimation techniques, in order to evaluate the accuracy of the proposed approach. The VB filter is further assessed on a data set of taste-response neural cells, showing that the proposed approach can effectively capture dynamical changes in neural responses in real time.

[1]  G. Kitagawa.  Theory and Methods , 1998 .

[2]  Emery N. Brown,et al.  A Bayesian statistical analysis of behavioral facilitation associated with deep brain stimulation , 2009, Journal of Neuroscience Methods.

[3]  Arnaud Doucet,et al.  Sequential Monte Carlo Methods to Train Neural Network Models , 2000, Neural Computation.

[4]  Emery N. Brown,et al.  The Time-Rescaling Theorem and Its Application to Neural Spike Train Data Analysis , 2002, Neural Computation.

[5]  A. Aertsen,et al.  Spike synchronization and rate modulation differentially involved in motor cortical function. , 1997, Science.

[6]  G. Kitagawa.  A self-organizing state-space model , 1998 .

[7]  John A. Quinn,et al.  Factorial Switching Linear Dynamical Systems Applied to Physiological Condition Monitoring , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Geir Storvik,et al.  Particle filters for state-space models with the presence of unknown static parameters , 2002, IEEE Trans. Signal Process..

[9]  Emery N. Brown,et al.  Computational Neuroscience: A Comprehensive Approach , 2022 .

[10]  Mahesan Niranjan,et al.  Estimating a State-Space Model from Point Process Observations: A Note on Convergence , 2010, Neural Computation.

[11]  Thomas Kailath,et al.  Nonlinear filtering with counting observations , 1975, IEEE Trans. Inf. Theory.

[12]  L. Fahrmeir,et al.  Dynamic Stochastic Models for Time-Dependent Ordered Paired Comparison Systems , 1994 .

[13]  E. Brown,et al.  A point-process model of human heartbeat intervals: new definitions of heart rate and heart rate variability. , 2005, American journal of physiology. Heart and circulatory physiology.

[14]  V. Šmídl,et al.  The Variational Bayes Method in Signal Processing , 2005 .

[15]  Zoubin Ghahramani,et al.  A Unifying Review of Linear Gaussian Models , 1999, Neural Computation.

[16]  J. Geweke,et al.  Bayesian estimation of state-space models using the Metropolis-Hastings algorithm within Gibbs sampling , 2001 .

[17]  Karl J. Friston,et al.  Variational free energy and the Laplace approximation , 2007, NeuroImage.

[18]  Shlomo Havlin,et al.  Scaling behaviour of heartbeat intervals obtained by wavelet-based time-series analysis , 1996, Nature.

[19]  D. J. MacGregor,et al.  A new method of spike modelling and interval analysis , 2009, Journal of Neuroscience Methods.

[20]  Matthew A. Wilson,et al.  Construction of Point Process Adaptive Filter Algorithms for Neural Systems Using Sequential Monte Carlo Methods , 2007, IEEE Transactions on Biomedical Engineering.

[21]  Matthew J. Beal.  Variational algorithms for approximate Bayesian inference , 2003 .

[22]  Emery N. Brown,et al.  Estimating a State-space Model from Point Process Observations Emery N. Brown , 2022 .

[23]  Robert J. Elliott,et al.  Discrete time filters for doubly stochastic poisson processes and other exponential noise models , 1999 .

[24]  J. Victor,et al.  Taste response variability and temporal coding in the nucleus of the solitary tract of the rat. , 2003, Journal of neurophysiology.

[25]  Hagai Attias,et al.  Inferring Parameters and Structure of Latent Variable Models by Variational Bayes , 1999, UAI.

[26]  Eric A. Wan,et al.  Dual Extended Kalman Filter Methods , 2002 .

[27]  Jonathan D Victor,et al.  Variability in responses and temporal coding of tastants of similar quality in the nucleus of the solitary tract of the rat. , 2008, Journal of neurophysiology.

[28]  Vaclav Edvard Benes,et al.  Recursive nonlinear estimation of a diffusion acting as the rate of an observed Poisson process , 1980, IEEE Trans. Inf. Theory.

[29]  Nando de Freitas,et al.  An Introduction to Sequential Monte Carlo Methods , 2001, Sequential Monte Carlo Methods in Practice.

[30]  Visakan Kadirkamanathan,et al.  Variational Estimation in Spatiotemporal Systems From Continuous and Point-Process Observations , 2012, IEEE Transactions on Signal Processing.

[31]  Mark A. Girolami,et al.  Markov Chain Monte Carlo Methods for State-Space Models with Point Process Observations , 2012, Neural Computation.

[32]  Masa-aki Sato,et al.  Online Model Selection Based on the Variational Bayes , 2001, Neural Computation.

[33]  S. Boll,et al.  Suppression of acoustic noise in speech using spectral subtraction , 1979 .

[34]  Wulfram Gerstner,et al.  A benchmark test for a quantitative assessment of simple neuron models , 2008, Journal of Neuroscience Methods.

[35]  R. Kohn,et al.  On Gibbs sampling for state space models , 1994 .

[36]  Emery N. Brown,et al.  Dynamic Analysis of Neural Encoding by Point Process Adaptive Filtering , 2004, Neural Computation.

[37]  Donald L. Snyder,et al.  Random Point Processes in Time and Space , 1991 .

[38]  Nando de Freitas,et al.  Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks , 2000, UAI.

[39]  'Unobserved' Monte Carlo method for identification of partially observed nonlinear state space systems. Part II. Counting process observations , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[40]  Visakan Kadirkamanathan,et al.  Point process modelling of the Afghan War Diary , 2012, Proceedings of the National Academy of Sciences.

[41]  Vaclav Smidl,et al.  The Restricted Variational Bayes Approximation in Bayesian Filtering , 2006, 2006 IEEE Nonlinear Statistical Signal Processing Workshop.