REVIEW 7 cited by
All-in-one simulation-based inference
Not yet reviewed by Pith; the record is open.
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
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.
Forward citations
Cited by 7 Pith papers
-
A Hierarchical Validity-Audit Framework for Neural Mass Models in Simulation-Based Inference: From Observational Coverage to Mechanistic Interpretation
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.
-
CoLT: The conditional localization test for assessing the accuracy of neural posterior estimates
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.
-
A COMPASS to Model Comparison and Simulation-Based Inference in Galactic Chemical Evolution
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.
-
Initial Luminally Deposited FGF4 Critically Influences Blastocyst Patterning
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...
-
Conformal C2ST: Turning weak classifiers into strong two-sample tests
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.
-
Position: The Future of Bayesian Prediction Is Prior-Fitted
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.
-
A Unified Framework for Simultaneous Parameter and Function Discovery in Differential Equations
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.
Discussion (0). Sign in to comment.