mlsanityData to 7 Oct 2026

Pareto Smoothed Importance Sampling

arXiv 1507.02646 ↗First posted 9 Jul 2015stat.CO stat.ME stat.MLcited by 42

Not on the map. Why?

Abstract (arXiv)

Importance weighting is a general way to adjust Monte Carlo integration to account for draws from the wrong distribution, but the resulting estimate can be highly variable when the importance ratios have a heavy right tail. This routinely occurs when there are aspects of the target distribution that are not well captured by the approximating distribution, in which case more stable estimates can be obtained by modifying extreme importance ratios. We present a new method for stabilizing importance weights using a generalized Pareto distribution fit to the upper tail of the distribution of the simulated importance ratios. The method, which empirically performs better than existing methods for stabilizing importance sampling estimates, includes stabilized effective sample size estimates, Monte Carlo error estimates, and convergence diagnostics. The presented Pareto hatk finite sample convergence rate diagnostic is useful for any Monte Carlo estimator.

Key facts. Cited by 42 papers in the map (data through 7 Oct 2026). First posted on arXiv 9 Jul 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 7; 2023: 5; 2024: 6; 2025: 11; 2026: 7.

42 papers.

Sort:

By area:

  1. 2021Variational Refinement for Importance Sampling Using the Forward Kullback-Leibler Divergencecited by 29
  2. 2021Pathfinder: Parallel quasi-Newton variational inferencecited by 26
  3. 2023Sensitivity-Aware Amortized Bayesian Inferencecited by 20
  4. 2021Bayesian hierarchical stacking: Some models are (somewhere) usefulcited by 14
  5. 2021q-Paths: Generalizing the Geometric Annealing Path using Power Meanscited by 12
  6. 2024Amortized Bayesian Multilevel Modelscited by 12
  7. 2024Variational Inference for Uncertainty Quantification: an Analysis of Trade-offscited by 11
  8. 2025Robust Amortized Bayesian Inference with Self-Consistency Losses on Unlabeled Datacited by 11
  9. 2024Variational Inference in Location-Scale Families: Exact Recovery of the Mean and Correlation Matrixcited by 9
  10. 2021The computational asymptotics of Gaussian variational inference and the Laplace approximationcited by 8
  11. 2026BayesFlow 2: Multi-Backend Amortized Bayesian Inference in Pythoncited by 7
  12. 2022Mathematical Theory of Bayesian Statistics for Unknown Information Sourcecited by 5
30 more
  1. 2022Meta-Uncertainty in Bayesian Model Comparisoncited by 5
  2. 2025Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimationcited by 5
  3. 2025Reinforced sequential Monte Carlo for amortised samplingcited by 5
  4. 2021Mitigating Statistical Bias within Differentially Private Synthetic Datacited by 4
  5. 2022Uphill Roads to Variational Tightness: Monotonicity and Monte Carlo Objectivescited by 3
  6. 2023Fast post-process Bayesian inference with Variational Sparse Bayesian Quadraturecited by 3
  7. 2024Demonstrating the power and flexibility of variational assumptions for amortized neural posterior estimation in environmental applicationscited by 3
  8. 2024Amortized Bayesian Workflowcited by 3
  9. 2025Amortized Bayesian Mixture Modelscited by 3
  10. 2025PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inferencecited by 3
  11. 2025Generalized Guarantees for Variational Inference in the Presence of Even and Elliptical Symmetrycited by 3
  12. 2022Efficient Bayes Inference in Neural Networks through Adaptive Importance Samplingcited by 2
  13. 2022Bayesian score calibration for approximate modelscited by 2
  14. 2023A Heavy-Tailed Algebra for Probabilistic Programmingcited by 2
  15. 2024Differentiable Pareto-Smoothed Weighting for High-Dimensional Heterogeneous Treatment Effect Estimationcited by 2
  16. 2025FlowVAT: Normalizing Flow Variational Inference with Affine-Invariant Temperingcited by 2
  17. 2025Normalizing Flow Regression for Bayesian Inference with Offline Likelihood Evaluationscited by 1
  18. 2026Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidancecited by 1
  19. 2022Robust leave-one-out cross-validation for high-dimensional Bayesian modelscited by 0
  20. 2022Autoencoded sparse Bayesian in-IRT factorization, calibration, and amortized inference for the Work Disability Functional Assessment Batterycited by 0
  21. 2023Locking and Quacking: Stacking Bayesian model predictions by log-pooling and superpositioncited by 0
  22. 2023Beyond Bayesian Model Averaging over Paths in Probabilistic Programs with Stochastic Supportcited by 0
  23. 2025Bayesian Ensembling: Insights from Online Optimization and Empirical Bayescited by 0
  24. 2025Generalized promotion time cure model: A new modeling framework to identify cell-type-specific genes and improve survival prognosiscited by 0
  25. 2025A Critical Review of Monte Carlo Algorithms Balancing Performance and Probabilistic Accuracy with AI Augmented Frameworkcited by 0
  26. 2026AI4BayesCode: From Natural Language Descriptions to Validated Modular Stateful Bayesian Samplerscited by 0
  27. 2026Beyond Effective Sample Size: Effective Number of Proposals for Adaptive Importance Samplingcited by 0
  28. 2026Correlation-Free Transition Path Sampling through Shooting Point Generation Guided by Committor Learningcited by 0
  29. 2026Large Language Bayes Is Not Reparameterisation-Invariantcited by 0
  30. 2026Agent Evaluation Reliability: More Tasks Won't (Always) Fix An Agent Leaderboardcited by 0