An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging

Sadjad Anzabi Zadeh; W. Nick Street; Barrett W Thomas

doi:10.48550/arxiv.2404.17187

Back

An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging

Preprint

Open access

An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging

Sadjad Anzabi Zadeh, W. Nick Street and Barrett W Thomas

ArXiv.org

Cornell University

04/26/2024

DOI: 10.48550/arxiv.2404.17187

Files and links (1)

url

https://doi.org/10.48550/arxiv.2404.17187View

Preprint (Author's original)This preprint has not been evaluated by subject experts through peer review. Preprints may undergo extensive changes and/or become peer-reviewed journal articles. Open Access

Abstract

Deep Reinforcement Learning is an effective tool for drug dosing for chronic condition management. However, the final protocol is generally a black box without any justification for its prescribed doses. This paper addresses this issue by proposing an explainable dosing protocol for warfarin using a Proximal Policy Optimization method combined with Policy Distillation. We introduce Action Forging as an effective tool to achieve explainability. Our focus is on the maintenance dosing protocol. Results show that the final model is as easy to understand and deploy as the current dosing protocols and outperforms the baseline dosing algorithms.

Computer Science - Learning

Details

Title: Subtitle: An Explainable Deep Reinforcement Learning Model for Warfarin Maintenance Dosing Using Policy Distillation and Action Forging
Creators: Sadjad Anzabi Zadeh
W. Nick Street
Barrett W Thomas
Resource Type: Preprint
Publication Details: ArXiv.org
DOI: 10.48550/arxiv.2404.17187
ISSN: 2331-8422
Publisher: Cornell University; Ithaca, New York
Language: English
Date posted: 04/26/2024
Academic Unit: Bus Admin College; Nursing; Computer Science; Business Analytics
Record Identifier: 9984621258502771

Metrics

22 Record Views