Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs
From MaRDI portal
Abstract: In this paper we propose an efficient stochastic optimization algorithm to search for Bayesian experimental designs such that the expected information gain is maximized. The gradient of the expected information gain with respect to experimental design parameters is given by a nested expectation, for which the standard Monte Carlo method using a fixed number of inner samples yields a biased estimator. In this paper, applying the idea of randomized multilevel Monte Carlo (MLMC) methods, we introduce an unbiased Monte Carlo estimator for the gradient of the expected information gain with finite expected squared -norm and finite expected computational cost per sample. Our unbiased estimator can be combined well with stochastic gradient descent algorithms, which results in our proposal of an optimization algorithm to search for an optimal Bayesian experimental design. Numerical experiments confirm that our proposed algorithm works well not only for a simple test problem but also for a more realistic pharmacokinetic problem.
Recommendations
- Nesterov-aided stochastic gradient methods using Laplace approximation for Bayesian design optimization
- Multilevel Monte Carlo estimation of expected information gains
- Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
- Stochastic optimization algorithms of a Bayesian design criterion for Bayesian parameter estimation of nonlinear regression models: Application in pharmacokinetics
- An induced natural selection heuristic for finding optimal Bayesian experimental designs
Cites work
- A Stochastic Approximation Method
- A tutorial on adaptive design optimization
- Adaptive subgradient methods for online learning and stochastic optimization
- Antithetic multilevel Monte Carlo estimation for multi-dimensional SDEs without Lévy area simulation
- Bayesian experimental design: A review
- Decision-making under uncertainty: using MLMC for efficient estimation of EVPPI
- Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
- Fast estimation of expected information gains for Bayesian experimental designs based on Laplace approximations
- General multilevel adaptations for stochastic approximation algorithms of Robbins-Monro and Polyak-Ruppert type
- GRADIENT-BASED STOCHASTIC OPTIMIZATION METHODS IN BAYESIAN EXPERIMENTAL DESIGN
- scientific article; zbMATH DE number 3850830 (Why is no real title available?)
- Inverse problems: a Bayesian perspective
- Lectures on Stochastic Programming
- Multi-level stochastic approximation algorithms
- Multilevel Monte Carlo estimation of expected information gains
- Multilevel Monte Carlo estimation of the expected value of sample information
- Multilevel Monte Carlo Path Simulation
- Multilevel simulation of functionals of Bernoulli random variables with application to basket credit derivatives
- Nesterov-aided stochastic gradient methods using Laplace approximation for Bayesian design optimization
- On a Measure of the Information Provided by an Experiment
- On the convergence of the Laplace approximation and noise-level-robustness of Laplace-based Monte Carlo methods for Bayesian inverse problems
- Stochastic simulation: Algorithms and analysis
- Towards Bayesian experimental design for nonlinear models that require a large number of sampling times
- Unbiased estimation with square root convergence for SDE models
Cited in
(11)- Unbiased estimation of the gradient of the log-likelihood in inverse problems
- Constructing unbiased gradient estimators with finite variance for conditional stochastic optimization
- Nesterov-aided stochastic gradient methods using Laplace approximation for Bayesian design optimization
- Multilevel Monte Carlo estimation of expected information gains
- Unbiased MLMC-based variational Bayes for likelihood-free inference
- Large-scale Bayesian optimal experimental design with derivative-informed projected neural network
- Modern Bayesian experimental design
- Optimal experimental design: formulations and computations
- Efficient estimation of expected information gain in Bayesian experimental design with multi-index Monte Carlo
- Bayesian experimental design for model discrepancy calibration: an auto-differentiable ensemble Kalman inversion approach
- Leveraging nested MLMC for sequential neural posterior estimation with intractable likelihoods
This page was built for publication: Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5028414)
