Reproduce “TestBED: An Open-Source Toolkit for Bayesian Experimental Design” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Gavin Kerrigan, Marcel Hedman, Sahel Iqbal, Tom Rainforth

The authors introduce TestBED, a toolkit that separates tasks, design strategies, objectives, and fitting algorithms for Bayesian experimental design. It lets researchers reuse or swap components and compare methods under shared infrastructure. In location-finding experiments, DAD’s Attention policy had an EIG upper bound of 14.359 ± 0.228 with explicit time input, versus 5.654 ± 1.061 without; this is one method and task.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.