Variational Bayesian inference for fMRI time series

By W. D. Penny, Stefan J. Kiebel, Karl J. Friston. 2003

207 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

Citations in the corpus

Constrained linear basis sets for HRF modelling using Variational Bayes (2004, methodology): “Penny et al. (2003) Using this approach we can approximate a posterior distribution p(hjy) with q(hjy) by minimising the KL-divergence, or equivalently by maximising the variational free energy, F, between them: F ¼ Z qðh j yÞlog pðy; hÞ qðh j yÞ dh ð18Þ To maximise this function, we need to ensure…”
Bayesian fMRI data analysis with sparse spatial basis function priors (2007, methodology): “We could, however, easily update the present model to deal with serial correlation using autoregressive (AR) processes as described in (Penny et al., 2003).”
Bayesian analysis of neuroimaging data in FSL (2009, background): “Minimising the difference between the true and approximate distributions results in tractable integrals when there are conjugate priors (Attias, 2000; Penny et al., 2003).”
Hierarchical Bayesian estimation for MEG inverse problem (2004, methodology): “The advantages of the VB estimate over the MAP estimate were also shown by Penny et al. (2003) for fMRI data analysis.”
Graph analysis of the human connectome: Promise, progress, and pitfalls (2013, unknown): “Using a variational Bayes approach, such an approach would then permit inversion of generative network models from empirical data and allow investigators to disambiguate between competing hypotheses that each embody — an approach that has been very successfully employed in other neuroimaging areas (”
Bayesian fMRI time series analysis with spatial priors (2005, methodology): “The posteriors over the noise precision and autoregressive coefficients are identical to those defined in previous work (Penny et al., 2003).”
Comparing Dynamic Causal Models using AIC, BIC and Free Energy (2012, background): “…2000; Gelman et al., 1995; Mackay, 2003), an approach that is now widely used in statistics (Hoeting et al., 1999), signal processing (Penny and Roberts, 2002), machine learning (Beal and Ghahramani, 2003), and neuroimaging (Friston et al., 2008; Penny et al., 2003; Trujillo-Barreto et al., 2004).”
Variational Bayesian inversion of the equivalent current dipole model in EEG/MEG (2008, methodology): “…Bayes provides a fast and efficient approximation to the necessary integrals and has been applied successfully to source imaging in M/EEG (Daunizeau et al., 2007; Sato et al., 2004) and other problems in functional imaging (Flandin and Penny, 2007; Penny et al., 2003; Woolrich and Behrens, 2006).”
Comparing Families of Dynamic Causal Models (2010, methodology): “Mathematical models of scientific data can be formally compared using Bayesian model evidence [1–3], an approach that is now widely used in statistics [4], signal processing [5], machine learning [6], natural language processing [7], and neuroimaging [8–10].”
A Bayesian spatiotemporal model for very large data sets (2010, unknown): “Here we follow the work of Penny et al. (2007) who used joint GMRF priors to spatially regularize the GLM and autoregressive (AR) coefficients (GLM-AR model) (Penny et al., 2003).”