publications

2025

1. arXiv

   

   Simulation-Based Inference: A Practical Guide

   Michael Deistler^*, Jan Boelts^*, Peter Steinbach, and 11 more authors

   Aug 2025

   * equal contribution

   Abs DOI arXiv HTML PDF Code

   A central challenge in many areas of science and engineering is to identify model parameters that are consistent with prior knowledge and empirical data. Bayesian inference offers a principled framework for this task, but can be computationally prohibitive when models are defined by stochastic simulators. Simulation-based Inference (SBI) is a suite of methods developed to overcome this limitation, which has enabled scientific discoveries in fields such as particle physics, astrophysics, and neuroscience. The core idea of SBI is to train neural networks on data generated by a simulator, without requiring access to likelihood evaluations. Once trained, inference is amortized: the neural network can rapidly perform Bayesian inference on empirical observations without requiring additional training or simulations. In this tutorial, we provide a practical guide for practitioners aiming to apply SBI methods. We outline a structured SBI workflow and offer practical guidelines and diagnostic tools for every stage of the process—from setting up the simulator and prior, choosing and training inference networks, to performing inference and validating the results. We illustrate these steps through examples from astrophysics, psychophysics, and neuroscience.

2. JOSS

   sbi reloaded: A toolkit for simulation-based inference workflows

   Jan Boelts^*, Michael Deistler^*, Manuel Gloeckler, and 30 more authors

   Journal of Open Source Software, Apr 2025

   * equal contribution

   Abs DOI HTML PDF Code Website

   The sbi package provides a unified Python interface for simulation-based inference (SBI) algorithms, enabling researchers to perform Bayesian inference on models with intractable likelihoods across diverse scientific domains. This paper presents sbi reloaded, a comprehensive update that has evolved through extensive community contributions and user feedback. The package now includes multiple state-of-the-art inference algorithms (NPE, NLE, NRE, FMPE, MNLE, and more), enhanced diagnostics and validation tools, improved computational efficiency, and a flexible API supporting custom neural network architectures and proposal distributions. Key improvements include GPU acceleration, distributed training support, comprehensive testing infrastructure, and extensive documentation with tutorials. With over 30 contributors and a growing user base, sbi has become the standard toolkit for neural simulation-based inference, balancing ease of use with the flexibility needed for cutting-edge research.

2024

1. bioRxiv

   Dissecting origins of wiring specificity in dense cortical connectomes

   Philipp Harth, Daniel Udvary, Jan Boelts, and 4 more authors

   bioRxiv, Dec 2024

   Abs DOI HTML PDF

   The organization of synaptic connections between neurons shapes the computations performed by cortical circuits. Understanding the principles governing this organization requires dissecting the contributions of different factors to synaptic specificity. Here, we analyze dense electron microscopy reconstructions of cortical tissue to identify the origins of wiring specificity. We develop a computational framework that separates the contributions of neuronal morphology, spatial proximity, and cell-type-specific connectivity rules to the observed connectivity patterns. Our analysis reveals that while spatial proximity and morphological constraints explain a substantial fraction of connectivity, cell-type-specific rules add critical specificity that cannot be explained by geometry alone.

2023

1. PLOS CB

   

   Simulation-based inference for efficient identification of generative models in connectomics

   Jan Boelts, Philipp Harth, Richard Gao, and 6 more authors

   PLOS Computational Biology, Aug 2023

   Abs DOI HTML PDF Code

   Understanding the organizational principles of neural circuits requires identifying the rules governing synaptic connectivity between neurons. However, inferring these rules from experimental data is challenging due to the complexity of biological networks and limitations in data acquisition. Here, we present a simulation-based inference approach for identifying generative models of connectivity in anatomically detailed cortical circuits. Our method combines biologically realistic simulations of cortical connectivity with neural density estimation to perform Bayesian inference on connectivity parameters. We validate our approach on synthetic data and apply it to a dense reconstruction of rat barrel cortex, revealing distance-dependent connectivity rules and cell-type-specific biases. The method is computationally efficient, requiring orders of magnitude fewer simulations than traditional approaches.

2022

1. eLife

   

   Flexible and efficient simulation-based inference for models of decision-making

   Jan Boelts, Jan-Matthis Lueckmann, Richard Gao, and 1 more author

   eLife, 2022

   Abs DOI HTML PDF Code

   Inferring parameters of computational models that capture experimental data is a central task in cognitive neuroscience. Bayesian statistical inference methods usually require the ability to evaluate the likelihood of the model—however, for many models of interest in cognitive neuroscience, the associated likelihoods cannot be computed efficiently. Simulation-based inference (SBI) offers a solution to this problem by only requiring access to simulations produced by the model. Here, we provide a new SBI method that is substantially more simulation-efficient. Our approach, Mixed Neural Likelihood Estimation (MNLE), trains neural density estimators on model simulations to emulate the simulator, and is designed to capture both the continuous (e.g., reaction times) and discrete (choices) data of decision-making models. We demonstrate MNLE on two variants of the drift-diffusion model (DDM) and show that it is substantially more efficient than LANs, achieving similar likelihood accuracy with six orders of magnitude fewer training simulations.

2. ICLR

   GATSBI: Generative Adversarial Training for Simulation-Based Inference

   Poornima Ramesh, Jan-Matthis Lueckmann, Jan Boelts, and 3 more authors

   In International Conference on Learning Representations (ICLR), 2022

   Abs HTML PDF Code

   Simulation-based inference (SBI) enables Bayesian inference for models where evaluating the likelihood is intractable but simulating from the model is possible. Many SBI methods approximate the posterior distribution using neural density estimators, but training these estimators requires a large number of simulations. We introduce GATSBI (Generative Adversarial Training for Simulation-Based Inference), which leverages generative adversarial networks (GANs) to improve the sample efficiency of posterior estimation. GATSBI trains a discriminator to distinguish between samples from the true posterior and samples from a learned generator, while simultaneously training the generator to approximate the posterior. We demonstrate GATSBI’s effectiveness on benchmark problems from neuroscience and epidemiology, showing that it achieves comparable or better accuracy than existing methods while requiring significantly fewer simulations.

2021

1. AISTATS

   Benchmarking Simulation-Based Inference

   Jan-Matthis Lueckmann, Jan Boelts, David S. Greenberg, and 2 more authors

   In Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS), 2021

   Abs HTML PDF Code Website

   Simulation-based inference (SBI) has emerged as a powerful framework for performing Bayesian inference when the likelihood function is intractable but simulations from the model are feasible. We present a comprehensive benchmark for SBI methods, including tasks with different properties: dimensionality of parameters and data, simulation budget, and requirements for amortization. Our benchmark includes reference posteriors obtained with expensive Monte Carlo methods, standardized implementations of competing algorithms, and metrics for accuracy and computational efficiency. We evaluate six SBI algorithms across ten benchmark tasks, revealing strengths and weaknesses of different approaches.

2020

1. JOSS

   sbi: A toolkit for simulation-based inference

   Álvaro Tejero-Cantero^*, Jan Boelts^*, Michael Deistler^*, and 5 more authors

   Journal of Open Source Software, 2020

   * equal contribution

   Abs DOI HTML PDF Code Website

   Scientists and engineers use numerical simulators to model empirical phenomena. When the simulator is stochastic, or the phenomena under investigation are corrupted by noise, the simulation outputs and the empirical observations will not match exactly. In such cases, a key task is to infer the simulator parameters that best describe the observed data. When the likelihood of the simulator is intractable, conventional Bayesian inference methods are inapplicable. Simulation-based inference (SBI) provides a solution by using neural networks to learn the relationship between parameters and data from simulations. The sbi package implements recent SBI algorithms in a user-friendly Python API. It provides a unified interface for different algorithms, enabling users to easily switch between methods and compare their performance.