Variational log‐Gaussian point‐process methods for grid cells

We present practical solutions to applying Gaussian‐process (GP) methods to calculate spatial statistics for grid cells in large environments. GPs are a data efficient approach to inferring neural tuning as a function of time, space, and other variables. We discuss how to design appropriate kernels...

Full description

Saved in:
Bibliographic Details
Published in:Hippocampus 2023-12, Vol.33 (12), p.1235-1251
Main Authors: Rule, Michael Everett, Chaudhuri‐Vayalambrone, Prannoy, Krstulovic, Marino, Bauza, Marius, Krupic, Julija, O'Leary, Timothy
Format: Article
Language:English
Subjects:
Bayes Theorem
Bayesian analysis
Frequency dependence
Gaussian process
Grid Cells
Mathematical models
Normal Distribution
point process
spatial statistics
variational Bayesian inference
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present practical solutions to applying Gaussian‐process (GP) methods to calculate spatial statistics for grid cells in large environments. GPs are a data efficient approach to inferring neural tuning as a function of time, space, and other variables. We discuss how to design appropriate kernels for grid cells, and show that a variational Bayesian approach to log‐Gaussian Poisson models can be calculated quickly. This class of models has closed‐form expressions for the evidence lower‐bound, and can be estimated rapidly for certain parameterizations of the posterior covariance. We provide an implementation that operates in a low‐rank spatial frequency subspace for further acceleration, and demonstrate these methods on experimental data.
ISSN:1050-9631
1098-1063
DOI:10.1002/hipo.23577