mlsanityData to 7 Oct 2026

Approximate Bayesian Computation with Domain Expert in the Loop

arXiv 2201.12090 ↗First posted 28 Jan 2022stat.ML cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Approximate Bayesian computation (ABC) is a popular likelihood-free inference method for models with intractable likelihood functions. As ABC methods usually rely on comparing summary statistics of observed and simulated data, the choice of the statistics is crucial. This choice involves a trade-off between loss of information and dimensionality reduction, and is often determined based on domain knowledge. However, handcrafting and selecting suitable statistics is a laborious task involving multiple trial-and-error steps. In this work, we introduce an active learning method for ABC statistics selection which reduces the domain expert's work considerably. By involving the experts, we are able to handle misspecified models, unlike the existing dimension reduction methods. Moreover, empirical results show better posterior estimates than with existing methods, when the simulation budget is limited.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 13 of its 52 references that are in the map. First posted on arXiv 28 Jan 2022.

Builds on

13 of 52 references.

References · 13

  1. 2017Flexible statistical inference for mechanistic models of neural dynamicscited by 162
  2. 2017On Nesting Monte Carlo Estimatorscited by 82
  3. 2020Neural Approximate Sufficient Statistics for Implicit Modelscited by 72
  4. 2018Likelihood-free inference with emulator networkscited by 61
  5. 2016Fast ε-free Inference of Simulation Models with Bayesian Conditional Density Estimationcited by 50
  6. 2015Learning Summary Statistic for Approximate Bayesian Computation via Deep Neural Networkcited by 39
  7. 2020A General Method for Calibrating Stochastic Radio Channel Models with Kernelscited by 9
  8. 2017Improving drug sensitivity predictions in precision medicine through active expert knowledge elicitationcited by 8
  9. 2018ABC-CDE: Towards Approximate Bayesian Computation with Complex High-Dimensional Data and Limited Simulationscited by 8
  10. 2016Knowledge Elicitation via Sequential Probabilistic Inference for High-Dimensional Predictioncited by 5
  11. 2017User Modelling for Avoiding Overfitting in Interactive Knowledge Elicitation for Predictioncited by 4
  12. 2015Summary Statistics in Approximate Bayesian Computationcited by 3
1 more
  1. 2017Regression approaches for Approximate Bayesian Computationcited by 1

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 1; 2025: 5.

9 papers.

  1. 2023Learning Robust Statistics for Simulation-based Inference under Model Misspecificationcited by 48
  2. 2023Leveraging Self-Consistency for Data-Efficient Amortized Bayesian Inferencecited by 14
  3. 2023Optimally-Weighted Estimators of the Maximum Mean Discrepancy for Likelihood-Free Inferencecited by 13
  4. 2025Simulation-based Bayesian inference under model misspecificationcited by 9
  5. 2024Cost-aware simulation-based inferencecited by 6
  6. 2025Robust Simulation-Based Inference under Missing Data via Neural Processescited by 5
  7. 2025Multilevel neural simulation-based inferencecited by 5
  8. 2025Flow Matching Calibration for Simulation-Based Inference under Model Misspecificationcited by 1
  9. 2025Misspecification-robust amortised simulation-based inference using variational methodscited by 0