Targeted Active Learning for Preference-Based Treatment Effects on Multivariate Outcomes
Abstract
Treatment efficacy is traditionally demonstrated on the basis of a single primary outcome. However, clinical decision-making usually requires consideration of multiple outcomes, balancing expected benefits against potential risks. The relative value assigned to these outcomes varies substantially from one patient to another. Given a preference rule over outcome profiles, treatment effects and optimal policies can be defined and estimated. Such a rule is rarely available in practice: it must itself be estimated from pairwise comparisons of outcome profiles, which are costly to collect from clinical experts. We propose an active learning framework that selects which comparisons to query. Standard criteria maximize the information gained on the preference rule itself. We instead target the quantities of interest, and select the query that most reduces uncertainty on the treatment effect and on the optimal policy induced by the learned rule. Under a Gaussian process model of the preference rule, we derive a closed-form approximation of this criterion. On semi-synthetic data built from a Parkinson's disease cohort with 13 clinical outcomes, our criterion achieves lower treatment effect estimation error and lower policy regret than existing criteria at equal query budget.
Related
- Augmented Patient Preference Incorporated Reinforcement Learning (APP-RL) to Estimate the Optimal Dynamic Treatment Regime
- Chance-Constrained Personalized LLM Routing with Online Preference Learning
- Reliability-Weighted Direct Preference Optimization: Penalized Weights for Robust Preference Learning
- PREFER: Personalized Review Summarization with Online Preference Learning
- Learning in Continuous Games from Pairwise Preference Feedback
BibTeX
@misc{giordani2026targeted,
title = {{Targeted Active Learning for Preference-Based Treatment Effects on Multivariate Outcomes}},
author = {Lola Giordani and Mathieu Even and Chloé Geoffroy and Jean-Christophe Corvol and Raphaël Porcher and Federico Pavone},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.03824},
eprint = {2610.03824},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.03824},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Targeted Active Learning for Preference-Based Treatment Effects on Multivariate Outcomes
AU - Lola Giordani
AU - Mathieu Even
AU - Chloé Geoffroy
AU - Jean-Christophe Corvol
AU - Raphaël Porcher
AU - Federico Pavone
PY - 2026
DA - 2026/10/12
PB - arXiv
DO - 10.48550/arxiv.2610.03824
UR - https://arxiv.org/abs/2610.03824
AB - Treatment efficacy is traditionally demonstrated on the basis of a single primary outcome. However, clinical decision-making usually requires consideration of multiple outcomes, balancing expected benefits against potential risks. The relative value assigned to these outcomes varies substantially from one patient to another. Given a preference rule over outcome profiles, treatment effects and optimal policies can be defined and estimated. Such a rule is rarely available in practice: it must itself be estimated from pairwise comparisons of outcome profiles, which are costly to collect from clinical experts. We propose an active learning framework that selects which comparisons to query. Standard criteria maximize the information gained on the preference rule itself. We instead target the quantities of interest, and select the query that most reduces uncertainty on the treatment effect and on the optimal policy induced by the learned rule. Under a Gaussian process model of the preference rule, we derive a closed-form approximation of this criterion. On semi-synthetic data built from a Parkinson's disease cohort with 13 clinical outcomes, our criterion achieves lower treatment effect estimation error and lower policy regret than existing criteria at equal query budget.
ER -