Awesome Neural SBI

A community-sourced list of papers and resources on neural simulation-based inference, covering both methodological developments and domain applications. Given the nature of the field, the list is bound to be highly incomplete -- contributions are welcome!

Software and Resources
Papers: Methods
Papers: Application

Software and Resources

Code Packages and Benchmarks

sbi [Code] [Docs] [Paper]: General-purpose simulation-based inference toolkit.
sbibm [Code] [Docs] [Paper]: Simulation-based inference benchmarking framework.
swyft [Code] [Docs] [Paper]: Official implementation of Truncated Marginal Neural Ratio Estimation (TMNRE), a hyper-efficient, simulation-based inference technique for complex data and expensive simulators.
lampe [Code] [Docs]: Likelihood-free AMortized Posterior Estimation with PyTorch.
MadMiner [Code] [Docs] [Paper]: Machine learning–based inference toolkit for particle physics.
pydelfi [Code] [Docs] [Paper]: Early implementation of Density Estimation Likelihood-Free Inference (DELFI) with neural density estimators and adaptive acquisition of simulations.
carl [Code] [Docs] [Paper]: Early toolbox for neural network-based likelihood-free inference in Python.

Review Papers

The frontier of simulation-based inference [arXiv]
Kyle Cranmer, Johann Brehmer, Gilles Louppe

Discovery and Links

arXiv search for "simulation-based inference" or "likelihood-free inference"
Google Scholar search for "simulation-based inference" or "likelihood-free inference"
simulation-based-inference.org

Papers: Methods

Methodological and use-inspired papers. Listed in reverse-chronological order.

Sampling-Based Accuracy Testing of Posterior Estimators for General Inference [arXiv]
Pablo Lemos, Adam Coogan, Yashar Hezaveh, Laurence Perreault-Levasseur
Misspecification-robust Sequential Neural Likelihood [arXiv]
Ryan P. Kelly, David J. Nott, David T. Frazier, David J. Warne, Chris Drovandi
Monte Carlo Techniques for Addressing Large Errors and Missing Data in Simulation-based Inference [arXiv]
Bingjie Wang, Joel Leja, Ashley Villar, Joshua S. Speagle
Validation Diagnostics for SBI algorithms based on Normalizing Flows [arXiv]
Julia Linhart, Alexandre Gramfort, Pedro L. C. Rodrigues
Likelihood-free hypothesis testing [arXiv]
Patrik Róbert Gerber, Yury Polyanskiy
Maximum Likelihood Learning of Energy-Based Models for Simulation-Based Inference [arXiv]
Pierre Glaser, Michael Arbel, Arnaud Doucet, Arthur Gretton
Efficient identification of informative features in simulation-based inference [arXiv]
Jonas Beck, Michael Deistler, Yves Bernaerts, Jakob Macke, Philipp Berens
Robust Neural Posterior Estimation and Statistical Model Criticism [arXiv]
Daniel Ward, Patrick Cannon, Mark Beaumont, Matteo Fasiolo, Sebastian M Schmon
Contrastive Neural Ratio Estimation [arXiv]
Benjamin Kurt Miller, Christoph Weniger, Patrick Forré
Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models [arXiv]
Louis Sharrock, Jack Simons, Song Liu, Mark Beaumont
Truncated proposals for scalable and hassle-free simulation-based inference [arXiv]
Michael Deistler, Pedro J Goncalves, Jakob H Macke
New Machine Learning Techniques for Simulation-Based Inference: InferoStatic Nets, Kernel Score Estimation, and Kernel Likelihood Ratio Estimation [arXiv]
Kyoungchul Kong, Konstantin T. Matchev, Stephen Mrenna, Prasanth Shyamsundar
Score Modeling for Simulation-based Inference [arXiv]
Tomas Geffner, George Papamakarios, Andriy Mnih
Investigating the Impact of Model Misspecification in Neural Simulation-based Inference [arXiv]
Patrick Cannon, Daniel Ward, Sebastian M. Schmon
Towards Reliable Simulation-Based Inference with Balanced Neural Ratio Estimation [arXiv]
Arnaud Delaunoy, Joeri Hermans, François Rozet, Antoine Wehenkel, Gilles Louppe
Likelihood-Free Inference with Generative Neural Networks via Scoring Rule Minimization [arXiv]
Lorenzo Pacchiardi, Ritabrata Dutta
Simulation-Based Inference with Waldo: Confidence Regions by Leveraging Prediction Algorithms or Posterior Estimators for Inverse Problems [arXiv]
Luca Masserano, Tommaso Dorigo, Rafael Izbicki, Mikael Kuusela, Ann B. Lee
Learning Optimal Test Statistics in the Presence of Nuisance Parameters [arXiv]
Lukas Heinrich
GATSBI: Generative Adversarial Training for Simulation-Based Inference [arXiv]
Poornima Ramesh, Jan-Matthis Lueckmann, Jan Boelts, Álvaro Tejero-Cantero, David S. Greenberg, Pedro J. Gonçalves, Jakob H. Macke
Variational methods for simulation-based inference [arXiv]
Manuel Glöckler, Michael Deistler, Jakob H. Macke
Robust Bayesian Inference for Simulator-based Models via the MMD Posterior Bootstrap [arXiv]
Charita Dellaporta, Jeremias Knoblauch, Theodoros Damoulas, François-Xavier Briol
Flexible and efficient simulation-based inference for models of decision-making [bioRxiv]
Jan Boelts, Jan-Matthis Lueckmann, Richard Gao, Jakob H. Macke
Group equivariant neural posterior estimation [arXiv]
Maximilian Dax, Stephen R. Green, Jonathan Gair, Michael Deistler, Bernhard Schölkopf, Jakob H. Macke
A Trust Crisis In Simulation-Based Inference? Your Posterior Approximations Can Be Unfaithful [arXiv]
Joeri Hermans, Arnaud Delaunoy, François Rozet, Antoine Wehenkel, Volodimir Begy, Gilles Louppe
Arbitrary Marginal Neural Ratio Estimation for Simulation-based Inference [arXiv]
François Rozet, Gilles Louppe
Likelihood-Free Frequentist Inference: Confidence Sets with Correct Conditional Coverage [arXiv]
Niccolò Dalmasso, Luca Masserano, David Zhao, Rafael Izbicki, Ann B. Lee
Truncated Marginal Neural Ratio Estimation [arXiv] [Code]
Benjamin Kurt Miller, Alex Cole, Patrick Forré, Gilles Louppe, Christoph Weniger
MINIMALIST: Mutual INformatIon Maximization for Amortized Likelihood Inference from Sampled Trajectories [arXiv]
Giulio Isacchini, Natanael Spisak, Armita Nourmohammad, Thierry Mora, Aleksandra M. Walczak
Simulation-Based Inference with Approximately Correct Parameters via Maximum Entropy [arXiv]
Rainier Barrett, Mehrad Ansari, Gourab Ghoshal, Andrew D White
Sequential Neural Posterior and Likelihood Approximation [arXiv]
Samuel Wiqvist, Jes Frellsen, Umberto Picchini
Diagnostics for Conditional Density Models and Bayesian Inference Algorithms [arXiv]
David Zhao, Niccolò Dalmasso, Rafael Izbicki, Ann B. Lee
HNPE: Leveraging Global Parameters for Neural Posterior Estimation [arXiv]
Pedro L. C. Rodrigues, Thomas Moreau, Gilles Louppe, Alexandre Gramfort
Benchmarking Simulation-Based Inference [arXiv]
Jan-Matthis Lueckmann, Jan Boelts, David S. Greenberg, Pedro J. Gonçalves, Jakob H. Macke
Solving high-dimensional parameter inference: marginal posterior densities & Moment Networks [arXiv]
Niall Jeffrey, Benjamin D. Wandelt
Neural Empirical Bayes: Source Distribution Estimation and its Applications to Simulation-Based Inference [arXiv]
Maxime Vandegar, Michael Kagan, Antoine Wehenkel, Gilles Louppe
Neural Approximate Sufficient Statistics for Implicit Models [arXiv]
Yanzhi Chen, Dinghuai Zhang, Michael Gutmann, Aaron Courville, Zhanxing Zhu
Differentiable Likelihoods for Fast Inversion of 'Likelihood-Free' Dynamical Systems [arXiv]
Hans Kersting, Nicholas Krämer, Martin Schiegg, Christian Daniel, Michael Tiemann, Philipp Hennig
On Contrastive Learning for Likelihood-free Inference [arXiv]
Conor Durkan, Iain Murray, George Papamakarios
Automatic Posterior Transformation for Likelihood-Free Inference [arXiv]
David S. Greenberg, Marcel Nonnenmacher, Jakob H. Macke
Likelihood-free MCMC with Amortized Approximate Ratio Estimators [arXiv]
Joeri Hermans, Volodimir Begy, Gilles Louppe
Dynamic Likelihood-free Inference via Ratio Estimation (DIRE) [arXiv]
Traiko Dinev, Michael U. Gutmann
Likelihood-free inference with an improved cross-entropy estimator [arXiv]
Markus Stoye, Johann Brehmer, Gilles Louppe, Juan Pavez, Kyle Cranmer
Mining gold from implicit models to improve likelihood-free inference [arXiv] [Code]
Johann Brehmer, Gilles Louppe, Juan Pavez, Kyle Cranmer
Likelihood-free inference with emulator networks [arXiv]
Jan-Matthis Lueckmann, Giacomo Bassetto, Theofanis Karaletsos, Jakob H. Macke
Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows [arXiv] [Code]
George Papamakarios, David C. Sterratt, Iain Murray
A Guide to Constraining Effective Field Theories with Machine Learning [arXiv]
Johann Brehmer, Kyle Cranmer, Gilles Louppe, Juan Pavez
Fast ε-free Inference of Simulation Models with Bayesian Conditional Density Estimation [arXiv]
George Papamakarios, Iain Murray
Approximating Likelihood Ratios with Calibrated Discriminative Classifiers [arXiv]
Kyle Cranmer, Juan Pavez, Gilles Louppe

Papers: Application

Domain application of neural simulation-based inference. Papers listed in reverse-chronological order.

Cosmology, Astrophysics, and Astronomy

Neural posterior estimation for exoplanetary atmospheric retrieval [arXiv]
Malavika Vasist, François Rozet, Olivier Absil, Paul Mollière, Evert Nasedkin, Gilles Louppe
Debiasing Standard Siren Inference of the Hubble Constant with Marginal Neural Ratio Estimation [arXiv]
Samuel Gagnon-Hartman, John Ruan, Daryl Haggard
Calibrating cosmological simulations with implicit likelihood inference using galaxy growth observables [arXiv]
Yongseok Jo et al
DIGS: Deep Inference of Galaxy Spectra with Neural Posterior Estimation [arXiv]
Gourav Khullar, Brian Nord, Aleksandra Ciprijanovic, Jason Poh, Fei Xu
Detection is truncation: studying source populations with truncated marginal neural ratio estimation [arXiv]
Noemi Anau Montel, Christoph Weniger
SIMBIG : A Forward Modeling Approach To Analyzing Galaxy Clustering [arXiv]
ChangHoon Hahn et al
Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference [arXiv]
Maximilian Dax, Stephen R. Green, Jonathan Gair, Michael Pürrer, Jonas Wildberger, Jakob H. Macke, Alessandra Buonanno, Bernhard Schölkopf
One never walks alone: the effect of the perturber population on subhalo measurements in strong gravitational lenses [arXiv]
Adam Coogan, Noemi Anau Montel, Konstantin Karchev, Meiert W. Grootes, Francesco Nattino, Christoph Weniger
SICRET: Supernova Ia Cosmology with truncated marginal neural Ratio EsTimation [arXiv]
Konstantin Karchev, Roberto Trotta, Christoph Weniger
Inferring subhalo effective density slopes from strong lensing observations with neural likelihood-ratio estimation [arXiv]
Gemma Zhang, Siddharth Mishra-Sharma, Cora Dvorkin
Uncovering dark matter density profiles in dwarf galaxies with graph neural networks [arXiv]
Tri Nguyen, Siddharth Mishra-Sharma, Reuel Williams, Lina Necib
Estimating Cosmological Constraints from Galaxy Cluster Abundance using Simulation-Based Inference [arXiv]
Moonzarin Reza, Yuanyuan Zhang, Brian Nord, Jason Poh, Aleksandra Ciprijanovic, Louis Strigari
Towards reconstructing the halo clustering and halo mass function of N-body simulations using neural ratio estimation [arXiv]
Androniki Dimitriou, Christoph Weniger, Camila A. Correa
Estimating the warm dark matter mass from strong lensing images with truncated marginal neural ratio estimation [arXiv]
Noemi Anau Montel, Adam Coogan, Camila Correa, Konstantin Karchev, Christoph Weniger
Implicit Likelihood Inference of Reionization Parameters from the 21 cm Power Spectrum [arXiv]
Xiaosheng Zhao, Yi Mao, Benjamin D. Wandelt
Accelerated Bayesian SED Modeling using Amortized Neural Posterior Estimation [arXiv]
ChangHoon Hahn, Peter Melchior
Simulation-Based Inference of Strong Gravitational Lensing Parameters [arXiv]
Ronan Legin, Yashar Hezaveh, Laurence Perreault Levasseur, Benjamin Wandelt
Fast and Credible Likelihood-Free Cosmology with Truncated Marginal Neural Ratio Estimation [arXiv]
Alex Cole, Benjamin Kurt Miller, Samuel J. Witte, Maxwell X. Cai, Meiert W. Grootes, Francesco Nattino, Christoph Weniger
A neural simulation-based inference approach for characterizing the Galactic Center γ-ray excess [arXiv]
Siddharth Mishra-Sharma, Kyle Cranmer
Inferring dark matter substructure with astrometric lensing beyond the power spectrum [arXiv]
Siddharth Mishra-Sharma
Approximate Bayesian Neural Doppler Imaging [arXiv]
A. Asensio Ramos, C. Diaz Baso, O. Kochukhov
Lossless, Scalable Implicit Likelihood Inference for Cosmological Fields [arXiv]
T. Lucas Makinen, Tom Charnock, Justin Alsing, Benjamin D. Wandelt
Real-time gravitational-wave science with neural posterior estimation [arXiv]
Maximilian Dax, Stephen R. Green, Jonathan Gair, Jakob H. Macke, Alessandra Buonanno, Bernhard Schölkopf
Real-Time Likelihood-Free Inference of Roman Binary Microlensing Events with Amortized Neural Posterior Estimation [arXiv]
Keming Zhang, Joshua S. Bloom, B. Scott Gaudi, Francois Lanusse, Casey Lam, Jessica R. Lu
Towards constraining warm dark matter with stellar streams through neural simulation-based inference [arXiv]
Joeri Hermans, Nilanjan Banik, Christoph Weniger, Gianfranco Bertone, Gilles Louppe
Lightning-Fast Gravitational Wave Parameter Inference through Neural Amortization [arXiv]
Arnaud Delaunoy, Antoine Wehenkel, Tanja Hinderer, Samaya Nissanke, Christoph Weniger, Andrew R. Williamson, Gilles Louppe
The sum of the masses of the Milky Way and M31: a likelihood-free inference approach [arXiv]
Pablo Lemos, Niall Jeffrey, Lorne Whiteway, Ofer Lahav, Niam I Libeskind, Yehuda Hoffman
Likelihood-free inference with neural compression of DES SV weak lensing map statistics [arXiv]
Niall Jeffrey, Justin Alsing, François Lanusse
Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning [arXiv]
Johann Brehmer, Siddharth Mishra-Sharma, Joeri Hermans, Gilles Louppe, Kyle Cranmer
Fast likelihood-free cosmology with neural density estimators and active learning [arXiv]
Justin Alsing, Tom Charnock, Stephen Feeney, Benjamin Wandelt
Investigating the turbulent hot gas in X-COP galaxy clusters [arXiv]
Simon Dupourqué, Nicolas Clerc, Etienne Pointecouteau, Dominique Eckert, Stefano Ettori, Franco Vazza

Particle Physics

Simulation-based inference methods for particle physics [arXiv]
Johann Brehmer, Kyle Cranmer
MadMiner: Machine learning-based inference for particle physics [arXiv]
Johann Brehmer, Felix Kling, Irina Espejo, Kyle Cranmer
Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale [arXiv]
Atılım Güneş Baydin et al
Constraining Effective Field Theories with Machine Learning [arXiv]
Johann Brehmer, Kyle Cranmer, Gilles Louppe, Juan Pavez

Neuroscience

Simulation-based inference for efficient identification of generative models in computational connectomics [bioRxiv]
Jan Boelts, Philipp Harth, Richard Gao, Daniel Udvary, Felipe Yáñez, Daniel Baum, Hans-Christian Hege, Marcel Oberlaender, Jakob H. Macke
Likelihood approximation networks (LANs) for fast inference of simulation models in cognitive neuroscience [Paper]
Alexander Fengler, Lakshmi N Govindarajan, Tony Chen, Michael J Frank
Training deep neural density estimators to identify mechanistic models of neural dynamics [Paper]
Pedro J Gonçalves et al
Amortized Bayesian Inference for Models of Cognition [arXiv]
Stefan T. Radev, Andreas Voss, Eva Marie Wieschen, Paul-Christian Bürkner

Health and Medicine

Mutation rate, selection, and epistasis inferred from RNA virus haplotypes via neural posterior estimation [bioRxiv]
Itamar Caspi, Moran Meir, Nadav Ben Nun, Uri Yakhini, Adi Stern, Yoav Ram
Simulation-Based Inference for Whole-Brain Network Modeling of Epilepsy using Deep Neural Density Estimators [medRxiv]
Meysam Hashemi, Anirudh N. Vattikonda, Jayant Jha, Viktor Sip, Marmaduke M. Woodman, Fabrice Bartolomei, Viktor K. Jirsa
Simulation-Based Inference for Global Health Decisions [arXiv]
Christian Schroeder de Witt et al

Other Domains

Applications where multiple papers could not be grouped under a single heading.

Simulation-based inference of single-molecule force spectroscopy [arXiv]
Lars Dingeldein, Pilar Cossio, Roberto Covino
Normalizing flows for likelihood-free inference with fusion simulations [Paper]
C S Furia, R M Churchill
Amortized Bayesian Inference of GISAXS Data with Normalizing Flows [arXiv]
Maksim Zhdanov, Lisa Randolph, Thomas Kluge, Motoaki Nakatsutsumi, Christian Gutt, Marina Ganeva, Nico Hoffmann
Optimal Design of Experiments for Simulation-Based Inference of Mechanistic Acyclic Biological Networks [arXiv]
Vincent Zaballa, Elliot Hui
Simulation-based Bayesian inference for multi-fingered robotic grasping [arXiv]
Norman Marlier, Olivier Brüls, Gilles Louppe
Simulation-based inference of evolutionary parameters from adaptation dynamics using neural networks [bioRxiv]
Grace Avecilla, Julie N. Chuong, Fangfei Li, Gavin Sherlock, David Gresham, Yoav Ram

Application to Real Data

Applications of neural simulation-based inference beyond synthetic data.

SIMBIG : A Forward Modeling Approach To Analyzing Galaxy Clustering [arXiv]
ChangHoon Hahn et al
A neural simulation-based inference approach for characterizing the Galactic Center γ-ray excess [arXiv]
Siddharth Mishra-Sharma, Kyle Cranmer
Towards constraining warm dark matter with stellar streams through neural simulation-based inference (Preliminary) [arXiv]
Joeri Hermans, Nilanjan Banik, Christoph Weniger, Gianfranco Bertone, Gilles Louppe
Likelihood-free inference with neural compression of DES SV weak lensing map statistics [arXiv]
Niall Jeffrey, Justin Alsing, François Lanusse

renecotyfanboy/awesome-neural-sbi