Pith. sign in

REVIEW 13 cited by

All-in-one simulation-based inference

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2404.09636 v3 pith:AT3EXBKL submitted 2024-04-15 cs.LG cs.AIstat.ML

Manuel Gloeckler , Michael Deistler , Christian Weilbach , Frank Wood , Jakob H. Macke This is my paper · ORCID
classification cs.LGcs.AIstat.ML
keywords inferenceamortizedbayesiandatasimformersimulation-basedcurrentmodel
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Amortized Bayesian inference trains neural networks to solve stochastic inference problems using model simulations, thereby making it possible to rapidly perform Bayesian inference for any newly observed data. However, current simulation-based amortized inference methods are simulation-hungry and inflexible: They require the specification of a fixed parametric prior, simulator, and inference tasks ahead of time. Here, we present a new amortized inference method -- the Simformer -- which overcomes these limitations. By training a probabilistic diffusion model with transformer architectures, the Simformer outperforms current state-of-the-art amortized inference approaches on benchmark tasks and is substantially more flexible: It can be applied to models with function-valued parameters, it can handle inference scenarios with missing or unstructured data, and it can sample arbitrary conditionals of the joint distribution of parameters and data, including both posterior and likelihood. We showcase the performance and flexibility of the Simformer on simulators from ecology, epidemiology, and neuroscience, and demonstrate that it opens up new possibilities and application domains for amortized Bayesian inference on simulation-based models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Hierarchical Validity-Audit Framework for Neural Mass Models in Simulation-Based Inference: From Observational Coverage to Mechanistic Interpretation

    q-bio.QM 2026-07 conditional novelty 6.0 of 10

    A hierarchical audit framework separates model-coverage failure, summary-induced information loss, target non-identifiability, and joint parameter compensation in neural-mass simulation-based inference.

  2. CoLT: The conditional localization test for assessing the accuracy of neural posterior estimates

    stat.ML 2025-07 conditional novelty 6.0 of 10

    CoLT learns to locate and measure the largest local mismatch between a neural posterior and the true posterior, turning it into a valid hypothesis test.

  3. A COMPASS to Model Comparison and Simulation-Based Inference in Galactic Chemical Evolution

    astro-ph.GA 2025-07 conditional novelty 6.0 of 10

    A diffusion-based simulation inference framework selects NuGrid AGB plus IllustrisTNG core-collapse yields as the best explanation for solar-type stellar abundances, and infers a steep IMF slope and high SN Ia normalization.

  4. Initial Luminally Deposited FGF4 Critically Influences Blastocyst Patterning

    q-bio.QM 2025-05 conditional novelty 6.0 of 10

    A spatial-stochastic model of the mouse blastocyst shows that a large initial deposit of FGF4 in the blastocoel, about 9,500 copies, is required to produce correct EPI-PRE patterning, with the blastocoel as source and...

  5. Solving Empirical Bayes via Transformers

    cs.LG 2025-02 conditional novelty 6.0 of 10

    A transformer pre-trained on synthetic Poisson data can beat the classical NPMLE estimator on several empirical Bayes tasks and run about 100x faster.

  6. Does Unsupervised Domain Adaptation Improve the Robustness of Amortized Bayesian Inference? A Systematic Evaluation

    stat.ML 2025-02 conditional novelty 6.0 of 10

    Aligning simulated and observed summaries improves neural Bayesian inference under likelihood misspecification but degrades it under prior misspecification.

  7. Can Transformers Learn Full Bayesian Inference in Context?

    cs.LG 2025-01 conditional novelty 6.0 of 10

    A transformer trained on synthetic data can output posterior samples for GLMs, factor analysis, and Gaussian mixtures in context, matching HMC quality and beating several VI baselines.

  8. Estimation of System Parameters Including Repeated Cross-Sectional Data through Emulator-Informed Deep Generative Model

    cs.LG 2024-12 conditional novelty 6.0 of 10

    EIDGM uses a HyperPINN emulator and Wasserstein GAN to estimate ODE parameter distributions from repeated cross-sectional data, with synthetic validation on three systems and an amyloid beta application.

  9. Conformal C2ST: Turning weak classifiers into strong two-sample tests

    stat.ML 2025-07 conditional novelty 5.0 of 10

    Conformal C2ST turns any classifier's ranking scores into exact or asymptotic p-values, giving a two-sample test whose validity does not depend on classifier quality and whose power tracks the classifier's AUC.

  10. Position: The Future of Bayesian Prediction Is Prior-Fitted

    cs.LG 2025-05 conditional novelty 4.0 of 10

    PFNs, which amortize Bayesian inference by training on datasets sampled from a prior, are likely to supersede MCMC and variational inference for most prediction tasks, the authors argue.

  11. A Unified Framework for Simultaneous Parameter and Function Discovery in Differential Equations

    cs.LG 2025-05 reject novelty 4.0 of 10

    The paper proves identifiability conditions for ODE inverse problems with one unknown constant and one unknown function, and adds approximate error bounds when data points are close but not identical.

  12. Simulation-Based Inference: A Practical Guide

    stat.ML 2025-08 accept novelty 3.0 of 10

    A practical tutorial for simulation-based inference, with a structured workflow, reusable code, and three worked examples validated by calibration diagnostics.

  13. sbi reloaded: a toolkit for simulation-based inference workflows

    cs.LG 2024-11 accept novelty 2.0 of 10

    sbi is an updated, feature-rich PyTorch toolkit for neural simulation-based inference, supporting many inference methods, samplers, and diagnostics.

Pith tools