mlsanityData to 7 Oct 2026

VISA: Variational Inference with Sequential Sample-Average Approximations

arXiv 2403.09429 ↗First posted 14 Mar 2024stat.ML cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

We present variational inference with sequential sample-average approximation (VISA), a method for approximate inference in computationally intensive models, such as those based on numerical simulations. VISA extends importance-weighted forward-KL variational inference by employing a sequence of sample-average approximations, which are considered valid inside a trust region. This makes it possible to reuse model evaluations across multiple gradient steps, thereby reducing computational cost. We perform experiments on high-dimensional Gaussians, Lotka-Volterra dynamics, and a Pickover attractor, which demonstrate that VISA can achieve comparable approximation accuracy to standard importance-weighted forward-KL variational inference with computational savings of a factor two or more for conservatively chosen learning rates.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 26 of its 46 references that are in the map. First posted on arXiv 14 Mar 2024.

Builds on

26 of 46 references.

References · 26

  1. 2013Auto-Encoding Variational Bayescited by 3,980
  2. 2015Importance Weighted Autoencoderscited by 454
  3. 2013Black Box Variational Inferencecited by 317
  4. 2014A Stochastic Quasi-Newton Method for Large-Scale Optimizationcited by 68
  5. 2014Reweighted Wake-Sleepcited by 66
  6. 2018An Introduction to Probabilistic Programmingcited by 62
  7. 2019Elements of Sequential Monte Carlocited by 60
  8. 2020Markovian Score Climbing: Variational Inference with KL(p||q)cited by 46
  9. 2018Doubly Reparameterized Gradient Estimators for Monte Carlo Objectivescited by 41
  10. 2017Covariances, Robustness, and Variational Bayescited by 39
  11. 2013Automated Variational Inference in Probabilistic Programmingcited by 31
  12. 2021Pathfinder: Parallel quasi-Newton variational inferencecited by 26
14 more
  1. 2023Black Box Variational Inference with a Deterministic Objective: Faster, More Accurate, and Even More Black Boxcited by 17
  2. 2015Linear Response Methods for Accurate Covariance Estimates from Mean Field Variational Bayescited by 14
  3. 2021Generalized Doubly Reparameterized Gradient Estimatorscited by 14
  4. 2021Nested Variational Inferencecited by 14
  5. 2015A trust-region method for stochastic variational inference with applications to streaming datacited by 9
  6. 2022Transport Score Climbing: Variational Inference Using Forward KL and Adaptive Neural Transportcited by 8
  7. 2017Fast Black-box Variational Inference through Stochastic Trust-Region Optimizationcited by 7
  8. 2019Amortized Population Gibbs Samplers with Neural Sufficient Statisticscited by 6
  9. 2021Learning Proposals for Probabilistic Programs with Inference Combinatorscited by 6
  10. 2023Sample Average Approximation for Black-Box VIcited by 6
  11. 2017Bayesian Optimization for Probabilistic Programscited by 3
  12. 2018Revisiting Reweighted Wake-Sleep for Models with Stochastic Control Flowcited by 3
  13. 2022Sparse Bayesian Lasso via a Variable-Coefficient ℓ1 Penaltycited by 2
  14. 2019Approximate Variational Inference Based on a Finite Sample of Gaussian Latent Variablescited by 1

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Control, Optimal Transport and Neural Differential Equations in Supervised Learningcited by 1