Search Results for author: Michael Deistler

Found 12 papers, 9 papers with code

All-in-one simulation-based inference

1 code implementation15 Apr 2024 Manuel Gloeckler, Michael Deistler, Christian Weilbach, Frank Wood, Jakob H. Macke

Amortized Bayesian inference trains neural networks to solve stochastic inference problems using model simulations, thereby making it possible to rapidly perform Bayesian inference for any newly observed data.

Bayesian Inference Epidemiology

Diffusion Tempering Improves Parameter Estimation with Probabilistic Integrators for Ordinary Differential Equations

1 code implementation19 Feb 2024 Jonas Beck, Nathanael Bosch, Michael Deistler, Kyra L. Kadhim, Jakob H. Macke, Philipp Hennig, Philipp Berens

Ordinary differential equations (ODEs) are widely used to describe dynamical systems in science, but identifying parameters that explain experimental measurements is challenging.

Amortized Bayesian Decision Making for simulation-based models

1 code implementation5 Dec 2023 Mila Gorecki, Jakob H. Macke, Michael Deistler

Simulation-based inference (SBI) provides a powerful framework for inferring posterior distributions of stochastic simulators in a wide range of domains.

Decision Making

Adversarial robustness of amortized Bayesian inference

1 code implementation24 May 2023 Manuel Glöckler, Michael Deistler, Jakob H. Macke

Bayesian inference usually requires running potentially costly inference procedures separately for every new observation.

Adversarial Robustness Bayesian Inference

Generalized Bayesian Inference for Scientific Simulators via Amortized Cost Estimation

no code implementations NeurIPS 2023 Richard Gao, Michael Deistler, Jakob H. Macke

Generalized Bayesian Inference (GBI) aims to robustify inference for (misspecified) simulator models, replacing the likelihood-function with a cost function that evaluates the goodness of parameters relative to data.

Bayesian Inference

Efficient identification of informative features in simulation-based inference

1 code implementation21 Oct 2022 Jonas Beck, Michael Deistler, Yves Bernaerts, Jakob Macke, Philipp Berens

To this end, many SBI methods employ a set of summary statistics or scientifically interpretable features to estimate a surrogate likelihood or posterior.

Bayesian Inference

Truncated proposals for scalable and hassle-free simulation-based inference

1 code implementation10 Oct 2022 Michael Deistler, Pedro J Goncalves, Jakob H Macke

TSNPE performs sequential inference with truncated proposals, sidestepping the optimisation issues of alternative approaches.

Variational methods for simulation-based inference

1 code implementation ICLR 2022 Manuel Glöckler, Michael Deistler, Jakob H. Macke

We present Sequential Neural Variational Inference (SNVI), an approach to perform Bayesian inference in models with intractable likelihoods.

Bayesian Inference Variational Inference

Group equivariant neural posterior estimation

1 code implementation ICLR 2022 Maximilian Dax, Stephen R. Green, Jonathan Gair, Michael Deistler, Bernhard Schölkopf, Jakob H. Macke

We here describe an alternative method to incorporate equivariances under joint transformations of parameters and data.

SBI -- A toolkit for simulation-based inference

no code implementations17 Jul 2020 Alvaro Tejero-Cantero, Jan Boelts, Michael Deistler, Jan-Matthis Lueckmann, Conor Durkan, Pedro J. Gonçalves, David S. Greenberg, Jakob H. Macke

$\texttt{sbi}$ facilitates inference on black-box simulators for practising scientists and engineers by providing a unified interface to state-of-the-art algorithms together with documentation and tutorials.

Bayesian Inference

Tactile Hallucinations on Artificial Skin Induced by Homeostasis in a Deep Boltzmann Machine

no code implementations25 Jun 2019 Michael Deistler, Yagmur Yener, Florian Bergner, Pablo Lanillos, Gordon Cheng

In this work, we investigate the generation of tactile hallucinations on biologically inspired, artificial skin.

Cannot find the paper you are looking for? You can Submit a new open access paper.