Pareto-guided reinforcement learning for multi-objective ADMET optimization in generative drug design

Simple item page
dc.contributor.authorNguyen, Hoang-Myen
dc.contributor.authorVu, Nguyet-Hangen
dc.contributor.authorLam, Hoang Thanhen
dc.contributor.authorNguyen, Hoang D.en
dc.contributor.funderResearch Irelanden
dc.contributor.funderEuropean Regional Development Funden
dc.date.accessioned2026-01-06T12:43:39Z
dc.date.available2026-01-06T12:43:39Z
dc.date.issued2025en
dc.description.abstractMulti-objective optimization is fundamental to early drug discovery, where improving one ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) property often degrades others. Existing generative approaches commonly rely on scalarized rewards or descriptor-based objectives, limiting their ability to capture complex pharmacokinetic trade-offs. We present RL-Pareto, a Pareto-guided reinforcement learning framework that directly optimizes predictor-driven ADMET objectives using a transformer-based SELFIES generator and a panel of LightGBM models. A compact reference Pareto set provides a dominance-based reward signal that preserves the structure of trade-offs while encouraging broad exploration. The framework scales flexibly to 1–22 simultaneous objectives without retraining and includes a natural-language interface that enables users to specify goals in plain text. In a benchmark involving simultaneous optimization of solubility and toxicity, RL-Pareto outperforms five strong baselines, PMMG, REINVENT, DrugEx-PCD, DrugEx-PTD, and GMD-MO-LSO, achieving 100% validity and novelty, strong diversity, and the highest hypervolume, reflecting the broadest Pareto-front expansion. RL-Pareto also reaches the best solubility and lowest toxicity extremes. These results highlight RL-Pareto with predictor-driven feedback as a principled, scalable, and practical approach for multi-objective molecular design.en
dc.description.sponsorshipResearch Ireland (12-RC-2289-P2)en
dc.description.statusPeer revieweden
dc.description.versionAccepted Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationNguyen, H.-M., Vu, N.-H., Lam, H. T. and Nguyen, H. D. (2025) 'Pareto-guided reinforcement learning for multi-objective ADMET optimization in generative drug design', NeurIPS 2025 Workshop: AI4Science, San Diego, California, United States, 6 December 2025.en
dc.identifier.endpage11en
dc.identifier.startpage1en
dc.identifier.urihttps://hdl.handle.net/10468/18367
dc.language.isoenen
dc.relation.ispartofThirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025), San Diego, CA, 2-7 December 2025en
dc.rights© 2025, the Authors. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.en
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.statusNot peer revieweden
dc.subjectMulti-objective optimizationen
dc.subjectADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity)en
dc.subjectEarly drug discoveryen
dc.subjectRL-Paretoen
dc.subjectPareto-guided reinforcement learning frameworken
dc.titlePareto-guided reinforcement learning for multi-objective ADMET optimization in generative drug designen
dc.typeConference itemen
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2025_NeurIPS_Drug_Discovery.pdf
Size:
2.45 MB
Format:
Adobe Portable Document Format
Description:
Accepted Version
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
2.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Computer Science - Conference Items
ADVANCE CRT - Conference Items
Insight Centre for Data Analytics - Conference Items