mlsanityData to 14 Oct 2026

Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design

arXiv 2507.14057 ↗First posted 18 Jul 2025stat.ML cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

We develop a semi-amortized, policy-based, approach to Bayesian experimental design (BED) called Stepwise Deep Adaptive Design (Step-DAD). Like existing, fully amortized, policy-based BED approaches, Step-DAD trains a design policy upfront before the experiment. However, rather than keeping this policy fixed, Step-DAD periodically updates it as data is gathered, refining it to the particular experimental instance. This test-time adaptation improves both the flexibility and the robustness of the design strategy compared with existing approaches. Empirically, Step-DAD consistently demonstrates superior decision-making and robustness compared with current state-of-the-art BED methods.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 37 of its 80 references that are in the map. First posted on arXiv 18 Jul 2025.

Builds on

37 of 80 references.

References · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  3. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,920
  4. 2017Deep Setscited by 1,421
  5. 2020MOPO: Model-based Offline Policy Optimizationcited by 636
  6. 2020MOReL : Model-Based Offline Reinforcement Learningcited by 590
  7. 2023Efficient Online Reinforcement Learning with Offline Datacited by 363
  8. 2019Monte Carlo Gradient Estimation in Machine Learningcited by 283
  9. 2022Online Decision Transformercited by 196
  10. 2022Hybrid RL: Using Both Offline and Online Data Can Make RL Efficientcited by 161
  11. 2023Modern Bayesian Experimental Designcited by 125
  12. 2019Variational Bayesian Optimal Experimental Designcited by 116
25 more
  1. 2021Deep Adaptive Design: Amortizing Sequential Bayesian Experimental Designcited by 106
  2. 2017On Nesting Monte Carlo Estimatorscited by 83
  3. 2016Likelihood-free inference by ratio estimationcited by 81
  4. 2021On Statistical Bias In Active Learning: How and When To Fix Itcited by 75
  5. 2013Hidden Parameter Markov Decision Processes: A Semiparametric Regression Approach for Discovering Latent Task Parametrizationscited by 70
  6. 2021Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoodscited by 61
  7. 2019A Unified Stochastic Gradient Approach to Designing Bayesian-Optimal Experimentscited by 58
  8. 2020Bayesian Experimental Design for Implicit Models by Mutual Information Neural Estimationcited by 56
  9. 2022Optimizing Sequential Experimental Design with Deep Reinforcement Learningcited by 46
  10. 2021Bayesian decision-making under misspecified priors with applications to meta-learningcited by 44
  11. 2018Efficient Bayesian Experimental Design for Implicit Modelscited by 34
  12. 2016Sequential Bayesian optimal experimental design via approximate dynamic programmingcited by 28
  13. 2021Gradient-based Bayesian Experimental Design for Implicit Models using Mutual Information Lower Boundscited by 27
  14. 2021Bayesian Sequential Optimal Experimental Design for Nonlinear Models Using Policy Gradient Reinforcement Learningcited by 24
  15. 2020Sequential Bayesian Experimental Design for Implicit Models via Mutual Informationcited by 23
  16. 2024Amortized Bayesian Experimental Design for Decision-Makingcited by 21
  17. 2022Policy-Based Bayesian Experimental Design for Non-Differentiable Implicit Modelscited by 16
  18. 2022Robust Expected Information Gain for Optimal Bayesian Experimental Design Using Ambiguity Setscited by 12
  19. 2024Nesting Particle Filters for Experimental Design in Dynamical Systemscited by 12
  20. 2024Bayesian Experimental Design via Contrastive Diffusionscited by 11
  21. 2018Bayesian design of experiments for intractable likelihood models using coupled auxiliary models and multivariate emulationcited by 10
  22. 2019Bayesian decision-theoretic design of experiments under an alternative modelcited by 8
  23. 2022Characterizing the robustness of Bayesian adaptive experimental designs to active learning biascited by 8
  24. 2024PASOA- PArticle baSed Bayesian Optimal Adaptive designcited by 5
  25. 2024Recursive Nested Filtering for Efficient Amortized Bayesian Experimental Designcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 9.

12 papers.

  1. 2025BED-LLM: Intelligent Information Gathering with LLMs and Bayesian Experimental Designcited by 24
  2. 2025JADAI: Jointly Amortizing Adaptive Design and Bayesian Inferencecited by 4
  3. 2025A Geometric Approach to Optimal Experimental Designcited by 3
  4. 2026Constrained Bayesian Experimental Design via Online Planningcited by 2
  5. 2026Efficient Adaptive Data Acquisition via Pretrained Belief Representationscited by 2
  6. 2026Bayesian Experimental Design via Score Matchingcited by 1
  7. 2026Sequential Design of Genetic Circuits Under Uncertainty With Reinforcement Learningcited by 0
  8. 2026Goal-driven Bayesian Optimal Experimental Design for Robust Decision-Making Under Model Uncertaintycited by 0
  9. 2026Can In-Context Learning Support Intrinsic Curiosity?cited by 0
  10. 2026FairBED: A Bayesian Experimental Design Approach to Gathering Fairer Datacited by 0
  11. 2026NeuronSifter: An Agent-in-Twin for Prioritizing Central Nervous System Interventions with World Action Modelscited by 0
  12. 2026Information-theoretic receding-horizon active learning of nonlinear dynamical systemscited by 0