Doubly Robust Stein-Kernelized Monte Carlo Estimator: Simultaneous Bias-Variance Reduction and Supercanonical Convergence
arXiv:2110.12131
Abstract
Standard Monte Carlo computation is widely known to exhibit a canonical square-root convergence speed in terms of sample size. Two recent techniques, one based on control variate and one on importance sampling, both derived from an integration of reproducing kernels and Stein's identity, have been proposed to reduce the error in Monte Carlo computation to supercanonical convergence. This paper presents a more general framework to encompass both techniques that is especially beneficial when the sample generator is biased and noise-corrupted. We show our general estimator, which we call the doubly robust Stein-kernelized estimator, outperforms both existing methods in terms of mean squared error rates across different scenarios. We also demonstrate the superior performance of our method via numerical examples.
References in corpus (8)
- Consistency and robustness of kernel-based regression in convex risk minimization
- Stein Variational Gradient Descent as Gradient Flow
- A Kernel Test of Goodness of Fit
- Stein Variational Policy Gradient
- Analysis of Kernel Mean Matching under Covariate Shift
- Stein Variational Gradient Descent Without Gradient
- The reproducing Stein kernel approach for post-hoc corrected sampling
- Robust Importance Weighting for Covariate Shift