Prompt-to-policy: Leveraging large language models to guide deep reinforcement learning in public health emergencies
Computer Methods and Programs in Biomedicine · 2026-05 · 279 · 109289
DOI: 10.1016/j.cmpb.2026.109289 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-05. Verfasst von Hao Yang und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Changchun University of Science and Technology, Changchun, 130022, Jilin, China. Electronic address: 20212…. Hauptthemen laut MEDLINE: Public Health, COVID-19, Emergencies, Deep Learning. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 59 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Rapid and effective decision-making is critical in public health emergencies, where resource allocation must balance multiple objectives under uncertain conditions. Traditional optimization methods often struggle with computational complexity and real-time, heterogeneous data. To address these challenges, we propose a hybrid intelligent agent combining an enhanced Double Deep Q-Network (D2QN-JDA) with large language models (LLMs). The D2QN-JDA improves learning stability and adaptability through joint state-action inputs, dynamic exploration rate, and adaptive reward normalization. The LLM component uses Retrieval-Augmented Generation (RAG) to integrate structured and unstructured data for real-time decision support. Experiments based on data from the Hong Kong COVID-19 outbreak show that the D2QN-JDA outperforms dynamic programming, greedy algorithms, genetic algorithms, and Q-learning, achieving reductions in cost. The LLM component also outperforms manual and regex methods in both single- and multi-point data extraction, enhancing accuracy, recall, F1 score, cost, and time. Our framework effectively addresses complex, multi-objective resource allocation in public health crises.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science and Technology, Changchun University of Science and Technology, Changchun, 130022, Jilin, China. Electronic address: 2021200130@mails.cust.edu.cn.
- School of Computer Science and Technology, Changchun University of Science and Technology, Changchun, 130022, Jilin, China; International Business School, Guangdong University of Finance & Economics, Guangzhou, 510320,…
- College of Computer Science and Technology, Jilin University, Changchun, 130012, Jilin, China. Electronic address: yangfy21@mails.jlu.edu.cn.
- Student Affairs Office, Changchun University, Changchun, 130022, Jilin, China. Electronic address: zxwluck@ccu.edu.cn.
- School of Computer Science and Engineering, Guangzhou Institute of Science and Technology, Guangzhou, 510540, Guangdong, China. Electronic address: 20170055@gzist.edu.cn.
- School of Management Science and Information Engineering, Jilin University of Finance and Economics, Changchun, 130117, Jilin, China. Electronic address: 13500907882@163.com.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Yang, H., Zhang, C., Yang, F., Zhang, X., Zhao, Y., & Sun, H. (2026). Prompt-to-policy: Leveraging large language models to guide deep reinforcement learning in public health emergencies. Computer Methods and Programs in Biomedicine, 279, 109289. https://doi.org/10.1016/j.cmpb.2026.109289
Vancouver
Yang H, Zhang C, Yang F, Zhang X, Zhao Y, Sun H. Prompt-to-policy: Leveraging large language models to guide deep reinforcement learning in public health emergencies. Computer Methods and Programs in Biomedicine. 2026;279:109289. doi:10.1016/j.cmpb.2026.109289
BibTeX
@article{Yang2026,
author = {Yang, Hao and Zhang, Chijun and Yang, Feiyang and Zhang, Xinwei and Zhao, Yugang and Sun, He},
title = {Prompt-to-policy: Leveraging large language models to guide deep reinforcement learning in public health emergencies},
journal = {Computer Methods and Programs in Biomedicine},
year = {2026},
volume = {279},
pages = {109289},
doi = {10.1016/j.cmpb.2026.109289}
}
EUVIMED