6 papers
JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference
Niels Bracher, Lars Kühmichel, Desi R. Ivanova +3
We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointl…
Bayesian Surrogate Training on Multiple Data Sources: A Hybrid Modeling Strategy
Philipp Reiser, Paul-Christian Bürkner, Anneli Guthke
Surrogate models are often used as computationally efficient approximations to complex simulation models, enabling tasks such as solving inverse problems, sensitivity analysis, and…
Amortized Bayesian Mixture Models
Šimon Kucharský, Paul Christian Bürkner
Finite mixtures are a broad class of models useful in scenarios where observed data is generated by multiple distinct processes but without explicit information about the responsib…
Uncertainty-Aware Surrogate-based Amortized Bayesian Inference for Computationally Expensive Models
Stefania Scheurer, Philipp Reiser, Tim Brünnette +3
Bayesian inference typically relies on a large number of model evaluations to estimate posterior distributions. Established methods like Markov Chain Monte Carlo (MCMC) and Amortiz…
Gaussian distributional structural equation models: A framework for modeling latent heteroscedasticity
Luna Fazio, Paul-Christian Bürkner
Accounting for the complexity of psychological theories requires methods that can predict not only changes in the means of latent variables -- such as personality factors, creativi…
DGP-LVM: Derivative Gaussian process latent variable models
Soham Mukherjee, Manfred Claassen, Paul-Christian Bürkner
We develop a framework for derivative Gaussian process latent variable models (DGP-LVMs) that can handle multi-dimensional output data using modified derivative covariance function…