English

JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference

Machine Learning 2025-12-30 v1 Artificial Intelligence Machine Learning

Abstract

We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointly amortizes Bayesian adaptive design and inference by training a policy, a history network, and an inference network end-to-end. The networks minimize a generic loss that aggregates incremental reductions in posterior error along experimental sequences. Inference networks are instantiated with diffusion-based posterior estimators that can approximate high-dimensional and multimodal posteriors at every experimental step. Across standard adaptive design benchmarks, JADAI achieves superior or competitive performance.

Keywords

Cite

@article{arxiv.2512.22999,
  title  = {JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference},
  author = {Niels Bracher and Lars Kühmichel and Desi R. Ivanova and Xavier Intes and Paul-Christian Bürkner and Stefan T. Radev},
  journal= {arXiv preprint arXiv:2512.22999},
  year   = {2025}
}
R2 v1 2026-07-01T08:43:33.076Z