Transformer-Based Reinforcement Learning for Phase I Dose Allocation
Abstract
Phase I dose-finding trials must learn from sparse toxicity observations while limiting exposure to overly toxic doses. We develop a single reinforcement-learning policy for trials with different numbers of ordered dose levels. Each dose is represented as a token containing its normalized position, accumulated patient count, observed dose-limiting toxicities (DLTs), and current-dose indicator. A Transformer encoder summarizes the dose tokens, trial-conditioned cross-attention produces a fixed-dimensional trial representation, and a three-action policy chooses de-escalation, staying, or escalation. The final maximum tolerated dose (MTD) decision is deliberately kept outside the learned policy: after the planned sample size, weighted isotonic regression and a prespecified safety cap select the dose closest to the target toxicity probability. In the simulations, toxicity-penalized policies reduce DLT burden while unpenalized policies improve selection accuracy at the cost of greater toxicity. These results demonstrate a controllable safety--accuracy trade-off but do not establish superiority over conventional dose-finding designs.
Related
- Efficient Joint Trajectory Optimization and Resource Allocation for Secure UAV Networks: A Deep Reinforcement Learning Approach
- Deep Reinforcement Learning-Aided Optimization of Multi-Interface Allocation for Short-Packet Communications
- Reinforcement learning-driven optimal capital allocation for intelligent finance
- Deep Reinforcement Learning for Optimization of STAR-RIS Phase and Energy Splitting Coefficients in OTFS-NOMA Framework
- Bayesian Optimization for Dose Finding with Two Agents: Participant Allocation and Final Selection
BibTeX
@misc{yuan2026transformerbased,
title = {{Transformer-Based Reinforcement Learning for Phase I Dose Allocation}},
author = {Shijie Yuan},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.00395},
eprint = {2611.00395},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.00395},
}
RIS
TY - UNPB
TI - Transformer-Based Reinforcement Learning for Phase I Dose Allocation
AU - Shijie Yuan
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.00395
UR - https://arxiv.org/abs/2611.00395
AB - Phase I dose-finding trials must learn from sparse toxicity observations while limiting exposure to overly toxic doses. We develop a single reinforcement-learning policy for trials with different numbers of ordered dose levels. Each dose is represented as a token containing its normalized position, accumulated patient count, observed dose-limiting toxicities (DLTs), and current-dose indicator. A Transformer encoder summarizes the dose tokens, trial-conditioned cross-attention produces a fixed-dimensional trial representation, and a three-action policy chooses de-escalation, staying, or escalation. The final maximum tolerated dose (MTD) decision is deliberately kept outside the learned policy: after the planned sample size, weighted isotonic regression and a prespecified safety cap select the dose closest to the target toxicity probability. In the simulations, toxicity-penalized policies reduce DLT burden while unpenalized policies improve selection accuracy at the cost of greater toxicity. These results demonstrate a controllable safety--accuracy trade-off but do not establish superiority over conventional dose-finding designs.
ER -