Variational Bayesian identification and prediction of stochastic nonlinear dynamic causal models.

Variational Bayesian identification and prediction of stochastic nonlinear dynamic causal models.
复制标题

DOI:
10.1016/j.physd.2009.08.002
复制
发表时间:
2009-11-01
期刊:
Physica D. Nonlinear phenomena
影响因子:
--
通讯作者:
Kiebel SJ
Kiebel SJ
中科院分区:
其他
文献类型:
--
作者:
Daunizeau J;Friston KJ;Kiebel SJ

文献摘要

参考文献

被引文献

相似文献

本文描述了一种用于非线性随机动力模型近似推理的通用变分贝叶斯方法。该方案扩展了已建立的隐状态近似推理,以涵盖:(i)非线性演化和观测函数,(ii)未知参数和(精度)超参数,以及(iii)不确定性下的模型比较和预测。模型识别或反演需要估计模型的边际似然或证据。这个困难的积分问题可以通过使用变分微积分的结果优化证据的自由能界来解决。这产生了一种确定性更新方案,该方案优化了对未知模型变量的后验密度的近似。我们在非线性随机动态层次模型的背景下推导了这种变分贝叶斯格式,用于模型识别和时间序列预测。该方案的计算复杂度与扩展卡尔曼滤波器相当,这在反演高维模型或长时间序列时至关重要。通过蒙特卡罗模拟,我们利用混沌动态系统的三个随机变量来评估这种变分贝叶斯方法的估计效率。我们还证明了该方法的模型比较能力、自一致性和预测能力。
In this paper, we describe a general variational Bayesian approach for approximate inference on nonlinear stochastic dynamic models. This scheme extends established approximate inference on hidden-states to cover: (i) nonlinear evolution and observation functions, (ii) unknown parameters and (precision) hyperparameters and (iii) model comparison and prediction under uncertainty. Model identification or inversion entails the estimation of the marginal likelihood or evidence of a model. This difficult integration problem can be finessed by optimising a free-energy bound on the evidence using results from variational calculus. This yields a deterministic update scheme that optimises an approximation to the posterior density on the unknown model variables. We derive such a variational Bayesian scheme in the context of nonlinear stochastic dynamic hierarchical models, for both model identification and time-series prediction. The computational complexity of the scheme is comparable to that of an extended Kalman filter, which is critical when inverting high dimensional models or long time-series. Using Monte-Carlo simulations, we assess the estimation efficiency of this variational Bayesian approach using three stochastic variants of chaotic dynamic systems. We also demonstrate the model comparison capabilities of the method, its self-consistency and its predictive power.
DOI: 10.1007/bf02530508
发表时间: 2003-06-01
影响因子: 1
作者:
Doucet, A;Tadic, VB
通讯作者: Tadic, VB
DOI: 10.1016/j.neuroimage.2006.08.035
发表时间: 2007-01-01
期刊: NEUROIMAGE
影响因子: 5.7
作者:
Friston, Karl J.;Mattout, Jeremie;Penny, Will
通讯作者: Penny, Will
DOI: 10.1016/j.physd.2004.03.020
发表时间: 2004-09-15
影响因子: 4
作者:
Judd, K;Smith, LA
通讯作者: Smith, LA
DOI: 10.1016/j.neuroimage.2007.02.046
发表时间: 2007-06-01
期刊: NEUROIMAGE
影响因子: 5.7
作者:
Kiebel, Stefan J.;Garrido, Marta I.;Friston, Karl J.
通讯作者: Friston, Karl J.
DOI: 10.1073/pnas.0308265100
发表时间: 2004-02-17
影响因子: 11.1
作者:
Angeli, D;Ferrell, JE;Sontag, ED
通讯作者: Sontag, ED