Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.15292 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908889421709312 |
|---|---|
| author | Gloeckler, Manuel Manzano-Patrón, J. P. Sotiropoulos, Stamatios N. Schröder, Cornelius Macke, Jakob H. |
| author_facet | Gloeckler, Manuel Manzano-Patrón, J. P. Sotiropoulos, Stamatios N. Schröder, Cornelius Macke, Jakob H. |
| contents | Simulation plays a central role in scientific discovery. In many applications, the bottleneck is no longer running a simulator; it is choosing among large families of plausible simulators, each corresponding to different forward models/hypotheses consistent with observations. Over large model families, classical Bayesian workflows for model selection are impractical. Furthermore, amortized model selection methods typically hard-code a fixed model prior or complexity penalty at training time, requiring users to commit to a particular parsimony assumption before seeing the data. We introduce PRISM, a simulation-based encoder-decoder that infers a joint posterior over both discrete model structures and associated continuous parameters, while enabling test-time control of model complexity via a tunable model prior that the network is conditioned on. We show that PRISM scales to families with combinatorially many (up to billions) of model instantiations on a synthetic symbolic regression task. As a scientific application, we evaluate PRISM on biophysical modeling for diffusion MRI data, showing the ability to perform model selection across several multi-compartment models, on both synthetic and in vivo neuroimaging data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15292 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Scalable Simulation-Based Model Inference with Test-Time Complexity Control Gloeckler, Manuel Manzano-Patrón, J. P. Sotiropoulos, Stamatios N. Schröder, Cornelius Macke, Jakob H. Machine Learning Artificial Intelligence Simulation plays a central role in scientific discovery. In many applications, the bottleneck is no longer running a simulator; it is choosing among large families of plausible simulators, each corresponding to different forward models/hypotheses consistent with observations. Over large model families, classical Bayesian workflows for model selection are impractical. Furthermore, amortized model selection methods typically hard-code a fixed model prior or complexity penalty at training time, requiring users to commit to a particular parsimony assumption before seeing the data. We introduce PRISM, a simulation-based encoder-decoder that infers a joint posterior over both discrete model structures and associated continuous parameters, while enabling test-time control of model complexity via a tunable model prior that the network is conditioned on. We show that PRISM scales to families with combinatorially many (up to billions) of model instantiations on a synthetic symbolic regression task. As a scientific application, we evaluate PRISM on biophysical modeling for diffusion MRI data, showing the ability to perform model selection across several multi-compartment models, on both synthetic and in vivo neuroimaging data. |
| title | Scalable Simulation-Based Model Inference with Test-Time Complexity Control |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.15292 |