arXiv Statistics ML

Neural composite likelihood estimation: simulation based inference for time series

Neural Composite Likelihood Estimation (NCLE) extends simulation‑based inference to high‑dimensional time series by partitioning long sequences into equal‑sized batches. For each batch, a neural network estimates the likelihood via conditional density estimation, and the product of these batch likelihoods forms an approximate composite likelihood. Frequentist inference is then performed by maximizing this composite likelihood to obtain a point estimate and by estimating the Godambe information matrix to derive confidence intervals.

arXiv Machine Learning
Sep 30

High-Dimensional Simulation-Based Inference in Latent Spaces

arXiv:2609.37381v1 Announce Type: new Abstract: Neural simulation-based inference (SBI) has been widely successful in inferring a relatively small number of interpretable parameters from potentially...

By Lars K\"uhmichel, Stefan T. Radev, Bhanu Prasanna Koppolu, Masoumeh Davoudi, Jerry M. Huang, Paul-Christian B\"urkner
arXiv Machine Learning
Sep 29

Mixed neural posterior estimation for simulators with discrete and continuous parameters

The paper extends Neural Posterior Estimation (NPE) to handle simulators whose parameter spaces contain both discrete and continuous dimensions. It introduces an inference network that factorizes the joint posterior into discrete and continuous components, using an autoregressive classifier for the discrete part and a generative model for the continuous part, trained jointly with a single simulation-based objective. A diagnostic tool for assessing calibration of the mixed posterior is also proposed, and the method is shown to produce accurate, calibrated posteriors on toy and real scientific simulators.

By Jan Boelts, Cornelius Schr\"oder, Jonas Beck, Jakob H. Macke, Michael Deistler, Daniel Gedon
arXiv Machine Learning
Oct 28

Adapting Neural Posterior Estimators for Simulation-Based Inference in Agent-Based Models

The paper presents a framework called SNPE‑TTT for adapting pre‑trained neural posterior estimators, specifically conditional normalizing flows, to new inference tasks in agent‑based models (ABMs). It explores four adaptation strategies—full‑parameter adaptation, low‑rank adaptation, and two gradient‑based update spaces—and evaluates them on economic simulation models with both tractable and intractable likelihoods. Results show that the framework can improve posterior recovery with a limited number of target simulations, though performance varies by task and adaptation method.

By Junyao Zhang, Jinglai Li, Junqi Tang
arXiv Machine Learning
Oct 16

Amortized Factor Inference Networks for Posterior Inference

The paper introduces Amortized Factor Inference Networks (AFINs), a single inference network that can generalize across varying priors, likelihoods, latent dimensions, and observation counts. AFINs represent a model as a list of typed prior and likelihood factors and use an encode–merge–decode architecture to produce a Gaussian posterior approximation, with learned parameters independent of latent dimensionality or observation count. Experiments on synthetic and real datasets demonstrate that a single trained AFIN achieves competitive posterior approximations at significantly lower test‑time cost than NUTS and per‑task variational inference, with optional importance sampling further improving accuracy.

By Joohwan Ko, Justin Domke
arXiv Statistics ML
Oct 12

SLICE: Sequential Likelihood-Free Inference via Conformal E-Values

SLICE: Sequential Likelihood-Free Inference via Conformal E-Values introduces a method for constructing exact anytime-valid confidence sequences for parameters of scientific models that can be simulated but have intractable likelihoods. The approach uses fresh simulations at each observation to conformalize a positive predictable score into a conditional e-value, whose product forms a test martingale that is inverted over the parameter space. The method is shown to be valid for any score, with the oracle likelihood-ratio score maximizing expected log growth, and is demonstrated on synthetic benchmarks, an audit of model routing in an LLM API, and microlensing data, where the resulting confidence sequences remain valid under continuous monitoring while neural scores remain competitive with exact likelihood ratios and application-specific scores.

By Stefano Cortinovis, Laura Battaglia, Fran\c{c}ois Caron