arXiv:2608.16466v2 Announce Type: replace-cross
Abstract: Bayesian optimal experimental design (BOED) aims to collect informative data by optimizing an expected utility reflecting the goals of an exp...
By David Chen, Michael Evans, Xinwei Li, Prateek Bansal, David J. Nott
TestBED is an open‑source toolkit that modularises Bayesian experimental design (BED) into reusable components such as tasks, objectives, training mechanisms, policy architectures, and evaluation procedures. It offers scalable implementations of core primitives like expected information gain estimators, enabling researchers to reproduce existing BED methods and explore new configurations within a common framework. The authors also use TestBED to run a large‑scale benchmark of contemporary BED approaches, highlighting critical methodological choices such as the role of time embeddings in design policy architectures.
By Gavin Kerrigan, Marcel Hedman, Sahel Iqbal, Tom Rainforth
GO-CBED is a goal‑oriented Bayesian framework for sequential causal experimental design that directly maximizes expected information gain on user‑specified causal quantities, rather than inferring the full causal model. It employs a variational lower bound estimator optimized with a transformer‑based policy network and normalizing flow posteriors, enabling real‑time, non‑myopic decision making. Experiments on synthetic structural causal models and semi‑synthetic gene regulatory networks show that GO‑CBED consistently outperforms existing baselines, especially under limited experimental budgets and complex causal mechanisms.
By Zheyu Zhang (University of Michigan), Jiayuan Dong (University of Michigan), Jie Liu (University of Michigan), Xun Huan (University of Michigan)
arXiv:2502. 08004v2 Announce Type: replace-cross Abstract: Simulation-based inference (SBI) is a method to perform inference on a variety of complex scientific models with challenging inference (inverse) problems.
By Vincent D. Zaballa, Elliot E. Hui
The paper introduces Generalised Bayesian Optimal Experimental Design (GBOED), extending generalised Bayesian inference to experimental design. It replaces the likelihood in Bayesian updates with a loss function, deriving a new acquisition function called Gibbs expected information gain. Empirical results show that GBOED improves robustness against outliers and mis-specified outcome noise distributions.
By Yasir Zubayr Barlas, Sabina J. Sloman, Samuel Kaski
The paper proposes a method to embed sequential Bayesian experimental design (BED) into large language models (LLMs) by training the models to reason about action selection using information-gain rewards. By specifying the experimental generative model separately from the LLM’s beliefs, the authors optimize expected information gain under a chosen model. Experiments on the Twenty Questions task show that the resulting policies outperform both inference- and training-time BED baselines while incurring lower computational costs, and the approach generalizes to unseen categories and transfers positively to the Medi-Q clinical dialogue benchmark.
By Anita Kriz, Antonin Berthon, Nicol\'as Astorga, Mihaela van der Schaar