Transformer-Based Reinforcement Learning Proposed for Phase I Dose Allocation
A new arXiv paper presents a single reinforcement learning approach using transformers to guide Phase I dose-finding trials, learning from sparse toxicity data while limiting patient exposure to overly toxic doses.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-03 05:00 · arXiv stat.ML
Transformer-Based Reinforcement Learning for Phase I Dose Allocation