# Characterizing the robustness of Bayesian adaptive experimental designs to active learning bias

arXiv: https://arxiv.org/abs/2205.13698 · HTML: https://mlsanity.com/paper/2205.13698/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 71 references that are in the map.
- First posted on arXiv 27 May 2022.

## Builds on (most cited first, 6 of 6)

- [On Statistical Bias In Active Learning: How and When To Fix It](https://mlsanity.com/paper/2101.11665/) (2021)
- [Feedback in Imitation Learning: The Three Regimes of Covariate Shift](https://mlsanity.com/paper/2102.02872/) (2021)
- [Misspecified Gaussian Process Bandit Optimization](https://mlsanity.com/paper/2111.05008/) (2021)
- [Sampling Bias in Deep Active Classification: An Empirical Study](https://mlsanity.com/paper/1909.09389/) (2019)
- [Sequential Bayesian optimal experimental design via approximate dynamic programming](https://mlsanity.com/paper/1604.08320/) (2016)
- [Addressing Bias in Active Learning with Depth Uncertainty Networks... or Not](https://mlsanity.com/paper/2112.06926/) (2021)

## Built on by (most cited first, 8 of 8)

- [Modern Bayesian Experimental Design](https://mlsanity.com/paper/2302.14545/) (2023)
- [Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design](https://mlsanity.com/paper/2507.14057/) (2025)
- [Metrics for Bayesian Optimal Experiment Design under Model Misspecification](https://mlsanity.com/paper/2304.07949/) (2023)
- [Bayesian Active Learning in the Presence of Nuisance Parameters](https://mlsanity.com/paper/2310.14968/) (2023)
- [Performance Comparisons of Reinforcement Learning Algorithms for Sequential Experimental Design](https://mlsanity.com/paper/2503.05905/) (2025)
- [Representative, Informative, and De-Amplifying: Requirements for Robust Bayesian Active Learning under Model Misspecification](https://mlsanity.com/paper/2506.07805/) (2025)
- [Active Learning Methods for Efficient Data Utilization and Model Performance Enhancement](https://mlsanity.com/paper/2504.16136/) (2025)
- [Supervised Learning as Lossy Compression: Characterizing Generalization and Sample Complexity via Finite Blocklength Analysis](https://mlsanity.com/paper/2602.04107/) (2026)

## Abstract (arXiv)

Bayesian adaptive experimental design is a form of active learning, which chooses samples to maximize the information they give about uncertain parameters. Prior work has shown that other forms of active learning can suffer from active learning bias, where unrepresentative sampling leads to inconsistent parameter estimates. We show that active learning bias can also afflict Bayesian adaptive experimental design, depending on model misspecification. We analyze the case of estimating a linear model, and show that worse misspecification implies more severe active learning bias. At the same time, model classes incorporating more "noise" - i.e., specifying higher inherent variance in observations - suffer less from active learning bias. Finally, we demonstrate empirically that insights from the linear model can predict the presence and degree of active learning bias in nonlinear contexts, namely in a (simulated) preference learning experiment.