Diffusion Models in Simulation-Based Inference: A Tutorial Review
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910004779417600 |
|---|---|
| author | Arruda, Jonas Bracher, Niels Köthe, Ullrich Hasenauer, Jan Radev, Stefan T. |
| author_facet | Arruda, Jonas Bracher, Niels Köthe, Ullrich Hasenauer, Jan Radev, Stefan T. |
| contents | Diffusion models have recently emerged as powerful learners for simulation-based inference (SBI), enabling fast and accurate estimation of latent parameters from simulated and real data. Their score-based formulation offers a flexible way to learn conditional or joint distributions over parameters and observations, thereby providing a versatile solution to various modeling problems. In this tutorial review, we synthesize recent developments on diffusion models for SBI, covering design choices for training, inference, and evaluation. We highlight opportunities created by various concepts such as guidance, score composition, flow matching, consistency models, and joint modeling. Furthermore, we discuss how efficiency and statistical accuracy are affected by noise schedules, parameterizations, and samplers. Finally, we illustrate these concepts with case studies across parameter dimensionalities, simulation budgets, and model types, and outline open questions for future research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_20685 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Diffusion Models in Simulation-Based Inference: A Tutorial Review Arruda, Jonas Bracher, Niels Köthe, Ullrich Hasenauer, Jan Radev, Stefan T. Machine Learning Methodology Diffusion models have recently emerged as powerful learners for simulation-based inference (SBI), enabling fast and accurate estimation of latent parameters from simulated and real data. Their score-based formulation offers a flexible way to learn conditional or joint distributions over parameters and observations, thereby providing a versatile solution to various modeling problems. In this tutorial review, we synthesize recent developments on diffusion models for SBI, covering design choices for training, inference, and evaluation. We highlight opportunities created by various concepts such as guidance, score composition, flow matching, consistency models, and joint modeling. Furthermore, we discuss how efficiency and statistical accuracy are affected by noise schedules, parameterizations, and samplers. Finally, we illustrate these concepts with case studies across parameter dimensionalities, simulation budgets, and model types, and outline open questions for future research. |
| title | Diffusion Models in Simulation-Based Inference: A Tutorial Review |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2512.20685 |