Automatic Laplace Collapsed Sampling: Scalable Marginalisation of Latent Parameters via Automatic Differentiation

TL;DR

ALCS introduces an automatic differentiation-based framework for scalable Bayesian marginalisation, combining Laplace approximation with nested sampling to efficiently explore hyperparameters in high-dimensional models.

Contribution

The paper presents a novel automatic differentiation approach that simplifies Bayesian evidence computation by collapsing latent variables, reducing computational complexity and enabling GPU parallelisation.

Findings

ALCS effectively marginalises high-dimensional latent variables in various models.

The method improves evidence estimates for heavy-tailed distributions using Student-t approximations.

A post-hoc ESS diagnostic localises failures in hyperparameter space without joint sampling.

Abstract

We present Automatic Laplace Collapsed Sampling (ALCS), a general framework for marginalising latent parameters in Bayesian models using automatic differentiation, which we combine with nested sampling to explore the hyperparameter space in a robust and efficient manner. At each nested sampling likelihood evaluation, ALCS collapses the high-dimensional latent variables $z$ to a scalar contribution via maximum a posteriori (MAP) optimisation and a Laplace approximation, both computed using autodiff. This reduces the effective dimension from $d_\theta + d_z$ to just $d_\theta$, making Bayesian evidence computation tractable for high-dimensional settings without hand-derived gradients or Hessians, and with minimal model-specific engineering. The MAP optimisation and Hessian evaluation are parallelised across live points on GPU-hardware, making the method practical at scale. We also show that automatic differentiation enables local approximations beyond Laplace to parametric families such as the Student-$t$, which improves evidence estimates for heavy-tailed latents. We validate ALCS on a suite of benchmarks spanning hierarchical, time-series, and discrete-likelihood models and establish where the Gaussian approximation holds. This enables a post-hoc ESS diagnostic that localises failures across hyperparameter space without expensive joint sampling.