Optimizing Social Utility in Sequential Experiments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Velasco, Ander Artola, Tsirtsis, Stratis, Gomez-Rodriguez, Manuel
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918488441880576
author Velasco, Ander Artola
Tsirtsis, Stratis
Gomez-Rodriguez, Manuel
author_facet Velasco, Ander Artola
Tsirtsis, Stratis
Gomez-Rodriguez, Manuel
contents Regulatory approval of products in high-stakes domains such as drug development requires statistical evidence of safety and efficacy through large-scale randomized controlled trials. However, the high financial cost of these trials may deter developers who lack absolute certainty in their product's efficacy, ultimately stifling the development of `moonshot' products that could offer high social utility. To address this inefficiency, in this paper, we introduce a statistical protocol for experimentation where the product developer (the agent) conducts a randomized controlled trial sequentially and the regulator (the principal) partially subsidizes its cost. By modeling the protocol using a belief Markov decision process, we show that the agent's optimal strategy can be found efficiently using dynamic programming. Further, we show that the social utility is a piecewise linear and convex function over the subsidy level the principal selects, and thus the socially optimal subsidy can also be found efficiently using divide-and-conquer. Simulation experiments using publicly available data on antibiotic development and approval demonstrate that our statistical protocol can be used to increase social utility by more than $35$$\%$ relative to standard, non-sequential protocols.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06520
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Optimizing Social Utility in Sequential Experiments
Velasco, Ander Artola
Tsirtsis, Stratis
Gomez-Rodriguez, Manuel
Computer Science and Game Theory
Machine Learning
Multiagent Systems
Methodology
Regulatory approval of products in high-stakes domains such as drug development requires statistical evidence of safety and efficacy through large-scale randomized controlled trials. However, the high financial cost of these trials may deter developers who lack absolute certainty in their product's efficacy, ultimately stifling the development of `moonshot' products that could offer high social utility. To address this inefficiency, in this paper, we introduce a statistical protocol for experimentation where the product developer (the agent) conducts a randomized controlled trial sequentially and the regulator (the principal) partially subsidizes its cost. By modeling the protocol using a belief Markov decision process, we show that the agent's optimal strategy can be found efficiently using dynamic programming. Further, we show that the social utility is a piecewise linear and convex function over the subsidy level the principal selects, and thus the socially optimal subsidy can also be found efficiently using divide-and-conquer. Simulation experiments using publicly available data on antibiotic development and approval demonstrate that our statistical protocol can be used to increase social utility by more than $35$$\%$ relative to standard, non-sequential protocols.
title Optimizing Social Utility in Sequential Experiments
topic Computer Science and Game Theory
Machine Learning
Multiagent Systems
Methodology
url https://arxiv.org/abs/2605.06520