Awesome Amortized Inference
+
+
Welcome to the Awesome Amortized Inference repository!
+This is a curated list of resources, including overviews, software, papers, and other resources related to amortized inference.
+Feel free to explore the entries below and use the provided BibTeX information for citation purposes.
+This is a community-driven project which is currently maintained by Marvin Schmitt.
+Contributions are always welcome, see CONTRIBUTING.md for a contribution guide.
This awesome list currently has some overlap with the awesome-neural-sbi list (Link) because
+amortized inference has gained populatity in the context of simulation-based inference (SBI) with neural networks.
+However, there is a trend towards broader amortized inference methods that are not necessarily simulation-based.
+This list aims to cover all amortized inference methods, including but not limited to simulation-based inference.
+We highly recommend checking out the awesome-neural-sbi list for more resources on modern simulation-based inference with neural networks.
++🚧 Under construction 🚧 +This repository is a development version under construction. The list of resources is just a small set for debugging and development purposes. +For now, we focus on developing features that we think could be useful for the community (e.g.,
+BibTeX-first structure). +Then, we will evaluate whether this project has the potential to be useful for the community in addition to existing lists +that are specifically designated to simulation-based inference and not amortized inference (which doesn't need to be simulation-based). +Once we figure out the direction and deem the project potentially useful for the community, we'll launch and add an extensive set of resources +-- hopefully with the help of many awesome people from the community 🧡
Contents
+ +Overview Articles
+-
+
-
+
Neural Methods for Amortized Inference (2024)
+
TLDR: Overview paper of amortized point estimators and full posterior estimators.
by Andrew Zammit-Mangion, Matthew Sainsbury-Dale, Raphaël Huser
[Paper]++Show BibTeX
++@misc{zammit-mangion2024neural, +title = {Neural Methods for Amortized Inference}, +publisher = {arXiv}, +year = {2024}, +author = {Zammit-Mangion, Andrew and Sainsbury-Dale, Matthew and Huser, Raphaël} +} ++
+ -
+
Normalizing flows for probabilistic modeling and inference (2021)
+
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by George Papamakarios, Eric Nalisnick, Danilo Jimenez Rezende, Shakir Mohamed, Balaji Lakshminarayanan
[Paper]++Show BibTeX
++@article{papamakarios2021normalizing, +title = {Normalizing flows for probabilistic modeling and inference}, +year = {2021}, +publisher = {JMLR.org}, +volume = {22}, +number = {1}, +issn = {1532-4435}, +journal = {J. Mach. Learn. Res.}, +month = {jan}, +articleno = {57}, +numpages = {64}, +author = {Papamakarios, George and Nalisnick, Eric and Rezende, Danilo Jimenez and Mohamed, Shakir and Lakshminarayanan, Balaji} +} ++
+ -
+
The frontier of simulation-based inference (2020)
+
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by Kyle Cranmer, Johann Brehmer, Gilles Louppe
[Paper]++Show BibTeX
++@article{Cranmer2020, +title = {The frontier of simulation-based inference}, +volume = {117}, +ISSN = {1091-6490}, +DOI = {10.1073/pnas.1912789117}, +number = {48}, +journal = {Proceedings of the National Academy of Sciences}, +publisher = {Proceedings of the National Academy of Sciences}, +year = {2020}, +pages = {30055-30062}, +author = {Cranmer, Kyle and Brehmer, Johann and Louppe, Gilles} +} ++
+
Software
+-
+
-
+
BayesFlow: Amortized Bayesian Workflows With Neural Networks
+
[Code]++Show BibTeX
++@article{radev2023bayesflow, +doi = {10.21105/joss.05702}, +year = {2023}, +publisher = {The Open Journal}, +volume = {8}, +number = {89}, +pages = {5702}, +title = {BayesFlow: Amortized Bayesian Workflows With Neural Networks}, +journal = {Journal of Open Source Software}, +author = {Radev, Stefan T. and Schmitt, Marvin and Schumacher, Lukas and Elsemüller, Lasse and Pratz, Valentin and Schälte, Yannik and Köthe, Ullrich and Bürkner, Paul-Christian} +} ++
+ -
+
sbi: A toolkit for simulation-based inference
+
[Code]++Show BibTeX
++@article{tejero-cantero2020sbi, +doi = {10.21105/joss.02505}, +year = {2020}, +publisher = {The Open Journal}, +volume = {5}, +number = {52}, +pages = {2505}, +title = {sbi: A toolkit for simulation-based inference}, +journal = {Journal of Open Source Software}, +author = {Tejero-Cantero, Alvaro and Boelts, Jan and Deistler, Michael and Lueckmann, Jan-Matthis and Durkan, Conor and Gonçalves, Pedro J. and Greenberg, David S. and Macke, Jakob H.} +} ++
+
Methodological Papers
+-
+
-
+
ASPIRE: Iterative Amortized Posterior Inference for Bayesian Inverse Problems (2024)
+
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by Rafael Orozco, Ali Siahkoohi, Mathias Louboutin, Felix J. Herrmann++Show BibTeX
++@misc{orozco2024aspire, +Title = {ASPIRE: Iterative Amortized Posterior Inference for Bayesian Inverse Problems}, +Year = {2024}, +Eprint = {arXiv:2405.05398}, +author = {Orozco, Rafael and Siahkoohi, Ali and Louboutin, Mathias and Herrmann, Felix J.} +} ++
+ -
+
Sensitivity-Aware Amortized Bayesian Inference (2024)
+
TLDR: Efficient amortized sensitivity analyses with context variables
by Lasse Elsemüller, Hans Olischläger, Marvin Schmitt, Paul-Christian Bürkner, Ullrich Koethe, Stefan T. Radev
[Code] [Paper]++Show BibTeX
++@article{elsemueller2024sensitivity, +title = {Sensitivity-Aware Amortized Bayesian Inference}, +journal = {Transactions on Machine Learning Research}, +issn = {2835-8856}, +year = {2024}, +author = {Elsem{\"u}ller, Lasse and Olischl{\"a}ger, Hans and Schmitt, Marvin and B{\"u}rkner, Paul-Christian and Koethe, Ullrich and Radev, Stefan T.} +} ++
+ -
+
Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference (2023)
+
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by Maximilian Dax, Stephen R. Green, Jonathan Gair, Michael Pürrer, Jonas Wildberger, Jakob H. Macke, Alessandra Buonanno, Bernhard Schölkopf
[Paper]++Show BibTeX
++@article{dax2023neural, +title = {Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference}, +volume = {130}, +ISSN = {1079-7114}, +DOI = {10.1103/physrevlett.130.171403}, +number = {17}, +journal = {Physical Review Letters}, +publisher = {American Physical Society (APS)}, +year = {2023}, +author = {Dax, Maximilian and Green, Stephen R. and Gair, Jonathan and P\"{u}rrer, Michael and Wildberger, Jonas and Macke, Jakob H. and Buonanno, Alessandra and Sch\"{o}lkopf, Bernhard} +} ++
+ -
+
JANA: Jointly Amortized Neural Approximation of Complex Bayesian Models (2023)
+
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by Stefan T. Radev, Marvin Schmitt, Valentin Pratz, Umberto Picchini, Ullrich Köthe, Paul-Christian Bürkner
[Paper]++Show BibTeX
++@inproceedings{radev2023jana, +title = {{JANA: Jointly Amortized Neural Approximation of Complex Bayesian Models}}, +booktitle = {Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence}, +pages = {1695--1706}, +year = {2023}, +volume = {216}, +series = {Proceedings of Machine Learning Research}, +publisher = {PMLR}, +author = {Radev, Stefan T. and Schmitt, Marvin and Pratz, Valentin and Picchini, Umberto and K\"othe, Ullrich and B\"urkner, Paul-Christian} +} ++
+
Application Papers
+-
+
- Evaluating Sparse Galaxy Simulations via Out-of-Distribution Detection and Amortized Bayesian Model Comparison (2024)
TLDR: Reading this paper? Please consider contributing a TLDR summary.
by Lingyi Zhou, Stefan T. Radev, William H. Oliver, Aura Obreja, Zehao Jin, Tobias Buck
[Paper]++Show BibTeX
++@inproceedings{zhou2024EvaluatingSparseGalaxy, +title = {Evaluating {{Sparse Galaxy Simulations}} via {{Out-of-Distribution Detection}} and {{Amortized Bayesian Model Comparison}}}, +booktitle = {38th {{Conference}} on {{Neural Information Processing Systems}}}, +year = {2024}, +author = {Zhou, Lingyi and Radev, Stefan T. and Oliver, William H. and Obreja, Aura and Jin, Zehao and Buck, Tobias} +} ++
+
Uncategorized
+-
+
- Flow Matching for Scalable Simulation-Based Inference (2023)
by Jonas Bernhard Wildberger, Maximilian Dax, Simon Buchholz, Stephen R Green, Jakob H. Macke, Bernhard Schölkopf
[Paper] [Code]++Show BibTeX
++@inproceedings{wildberger2023flow, +title = {Flow Matching for Scalable Simulation-Based Inference}, +booktitle = {Thirty-seventh Conference on Neural Information Processing Systems}, +year = {2023}, +url = {https://openreview.net/forum?id=D2cS6SoYlP}, +author = {Wildberger, Jonas Bernhard and Dax, Maximilian and Buchholz, Simon and Green, Stephen R and Macke, Jakob H. and Sch{\"o}lkopf, Bernhard} +} ++
+