Abstract:Abstract In generative modeling of neuroimaging data, such as dynamic causal modeling (DCM), one typically considers several alternative models, either to determine the most plausible explanation for observed data (Bayesian model selection) or to account for model uncertainty (Bayesian model averaging). Both procedures rest on estimates of the model evidence, a principled trade-off between model accuracy and complexity. In DCM, the log evidence is usually approximated using variational Bayes (VB) under the Laplace approximation (VBL). Although this approach is highly efficient, it makes distributional assumptions and can be vulnerable to local extrema. An alternative to VBL is Markov Chain Monte Carlo (MCMC) sampling, which is asymptotically exact but orders of magnitude slower than VB. This has so far prevented its routine use for DCM. This paper makes four contributions. First, we introduce a powerful MCMC scheme – thermodynamic integration (TI) – to neuroimaging and present a derivation that establishes a theoretical link to VB. Second, this derivation is based on a tutorial-like introduction to concepts of free energy in physics and statistics. Third, we present an implementation of TI for DCM that rests on population MCMC. Fourth, using simulations and empirical functional magnetic resonance imaging (fMRI) data, we compare log evidence estimates obtained by TI, VBL, and other MCMC-based estimators (prior arithmetic mean and posterior harmonic mean). We find that model comparison based on VBL gives reliable results in most cases, justifying its use in standard DCM for fMRI. Furthermore, we demonstrate that for complex and/or nonlinear models, TI may provide more robust estimates of the log evidence. Importantly, accurate estimates of the model evidence can be obtained with TI in acceptable computation time. This paves the way for using DCM in scenarios where the robustness of single-subject inference and model selection becomes paramount, such as differential diagnosis in clinical applications.

Target-aware Bayesian inference via generalized thermodynamic integration

Thermodynamic Bayesian Inference

An introduction to thermodynamic integration and application to dynamic causal models

The Controlled Thermodynamic Integral for Bayesian Model Evidence Evaluation

Thermodynamic Interpolation: A generative approach to molecular thermodynamics and kinetics

An adaptive scheduling scheme for calculating Bayes factors with thermodynamic integration using Simpson’s rule

Neural Thermodynamic Integration: Free Energies from Energy-based Diffusion Models

Generalized Posteriors in Approximate Bayesian Computation

Bayesian Parameter Estimation for Biochemical Reaction Networks Using Region-Based Adaptive Parallel Tempering

Bayesian Numerical Integration with Neural Networks

Thermodynamic integration for dynamic causal models

Efficient Bayesian estimation of the generalized Langevin equation from data

Boltzmann priors for Implicit Transfer Operators

Solving linear Bayesian inverse problems using a fractional total variation-Gaussian (FTG) prior and transport map

Generalized Laplace Approximation

A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit

Efficient, Multimodal, and Derivative-Free Bayesian Inference With Fisher-Rao Gradient Flows

Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations

Density Estimation via Binless Multidimensional Integration

Partitioned integrators for thermodynamic parameterization of neural networks