mlsanityData to 7 Oct 2026

Automated Efficient Estimation using Monte Carlo Efficient Influence Functions

arXiv 2403.00158 ↗First posted 29 Feb 2024stat.CO cs.LG stat.MEcited by 4

Not on the map. Why?

Abstract (arXiv)

Many practical problems involve estimating low dimensional statistical quantities with high-dimensional models and datasets. Several approaches address these estimation tasks based on the theory of influence functions, such as debiased/double ML or targeted minimum loss estimation. This paper introduces textitMonte Carlo Efficient Influence Functions (MC-EIF), a fully automated technique for approximating efficient influence functions that integrates seamlessly with existing differentiable probabilistic programming systems. MC-EIF automates efficient statistical estimation for a broad class of models and target functionals that would previously require rigorous custom analysis. We prove that MC-EIF is consistent, and that estimators using MC-EIF achieve optimal √N convergence rates. We show empirically that estimators using MC-EIF are at parity with estimators using analytic EIFs. Finally, we demonstrate a novel capstone example using MC-EIF for optimal portfolio selection.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 36 of its 75 references that are in the map. First posted on arXiv 29 Feb 2024.

Builds on

36 of 75 references.

References · 36

  1. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  2. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  3. 2015Automatic differentiation in machine learning: a surveycited by 846
  4. 2018Pyro: Deep Universal Probabilistic Programmingcited by 379
  5. 2019Why gradient clipping accelerates training: A theoretical justification for adaptivitycited by 335
  6. 2023Studying Large Language Model Generalization with Influence Functionscited by 311
  7. 2019Limitations of the Empirical Fisher Approximation for Natural Gradient Descentcited by 213
  8. 2019Orthogonal Statistical Learningcited by 150
  9. 2018Fast Approximate Natural Gradient Descent in a Kronecker-factored Eigenbasiscited by 143
  10. 2022If Influence Functions are the Answer, Then What is the Question?cited by 142
  11. 2016A Kronecker-factored approximate Fisher matrix for convolution layerscited by 138
  12. 2022Semiparametric doubly robust targeted double machine learning: a reviewcited by 129
24 more
  1. 2015Gradient Estimation Using Stochastic Computation Graphscited by 120
  2. 2019Exact Gaussian Processes on a Million Data Pointscited by 118
  3. 2017On Nesting Monte Carlo Estimatorscited by 82
  4. 2018A Swiss Army Infinitesimal Jackknifecited by 51
  5. 2021RieszNet and ForestRiesz: Automatic Debiased Machine Learning with Neural Nets and Random Forestscited by 42
  6. 2015The Influence Function of Semiparametric Estimatorscited by 36
  7. 2021Knowledge Distillation as Semiparametric Inferencecited by 29
  8. 2023Doubly Robust Self-Trainingcited by 18
  9. 2023The Fundamental Limits of Structure-Agnostic Functional Estimationcited by 12
  10. 2022ADEV: Sound Automatic Differentiation of Expected Values of Probabilistic Programscited by 11
  11. 2015Higher Order Tangent Spaces and Influence Functionscited by 9
  12. 2021Continuous-time targeted minimum loss-based estimation of intervention-specific mean outcomescited by 9
  13. 2019Covariance Matrix Estimation under Total Positivity for Portfolio Selectioncited by 8
  14. 2021Estimation of time-specific intervention effects on continuously distributed time-to-event outcomes by targeted maximum likelihood estimationcited by 7
  15. 2019The Variational Predictive Natural Gradientcited by 4
  16. 2019Estimating covariance and precision matrices along subspacescited by 4
  17. 2021Two-Stage TMLE to Reduce Bias and Improve Efficiency in Cluster Randomized Trialscited by 4
  18. 2022A Free Lunch with Influence Functions? Improving Neural Network Estimates with Concepts from Semiparametric Statisticscited by 4
  19. 2023Optimal randomized multilevel Monte Carlo for repeatedly nested expectationscited by 4
  20. 2023Targeted Maximum Likelihood Based Estimation for Longitudinal Mediation Analysiscited by 4
  21. 2022Efficient Targeted Learning of Heterogeneous Treatment Effects for Multiple Subgroupscited by 3
  22. 2023Targeted Learning on Variable Importance Measure for Heterogeneous Treatment Effectcited by 3
  23. 2022Data-Driven Influence Functions for Optimization-Based Causal Inferencecited by 1
  24. 2022Experiment-selector cross-validated targeted maximum likelihood estimator for hybrid RCT-external data studiescited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 1; 2026: 2.

4 papers.

  1. 2026EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environmentscited by 12
  2. 2024Debiasing Synthetic Data Generated by Deep Generative Modelscited by 2
  3. 2025Causal identification with Y0cited by 0
  4. 2026Kernel Debiased Plug-in Estimation based on the Universal Least Favorable Submodelcited by 0