Recursive Nested Filtering for Efficient Amortized Bayesian Experimental Design
Abstract
This paper introduces the Inside-Out Nested Particle Filter (IO-NPF), a novel fully recursive algorithm for amortized sequential Bayesian experimental design in the non-exchangeable setting. We frame policy optimization as maximum likelihood estimation in a non-Markovian state-space model, achieving (at most) $\mathcal{O}(T^2)$ computational complexity in the number of experiments. We provide theoretical convergence guarantees and introduce a backward sampling algorithm to reduce trajectory degeneracy. The IO-NPF offers a practical, extensible, and provably consistent approach to sequential Bayesian experimental design, demonstrating improved efficiency over existing methods.
Cite
Text
Iqbal et al. "Recursive Nested Filtering for Efficient Amortized Bayesian Experimental Design." NeurIPS 2024 Workshops: BDU, 2024.Markdown
[Iqbal et al. "Recursive Nested Filtering for Efficient Amortized Bayesian Experimental Design." NeurIPS 2024 Workshops: BDU, 2024.](https://mlanthology.org/neuripsw/2024/iqbal2024neuripsw-recursive/)BibTeX
@inproceedings{iqbal2024neuripsw-recursive,
title = {{Recursive Nested Filtering for Efficient Amortized Bayesian Experimental Design}},
author = {Iqbal, Sahel and Abdulsamad, Hany and Perez-Vieites, Sara and Särkkä, Simo and Corenflos, Adrien},
booktitle = {NeurIPS 2024 Workshops: BDU},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/iqbal2024neuripsw-recursive/}
}