Research map / Reinforcement learning / Visual and representation RL
Bootstrapping Personalized Insulin Therapy via Model-Based Reinforcement Learning: An In Silico Study
Sumana Basu, Flemming Kondrup, Adriana Romero-Soriano, Doina Precup
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use model-based H-step Deep Dyna-Q with transferred model instead of PID | insulin dosing policy · Adult#006, Simglucose | episode length: 23.17 -> 45.56 | Improved |
| use model-based H-step Deep Dyna-Q instead of basal-bolus | insulin dosing policy · Adolescent#010, Simglucose | time in range: 9.040 -> 17.69 | Improved |
| use model-based approach instead of PID | insulin dosing policy · Child#010, Simglucose | episode length: 16.44 -> 45.28 | Improved |
| match reference patient from the same insulin-sensitivity cluster (ablation) | model bootstrapping RL · Adult target, Figure 5 | policy performance | Improved |
| transfer model from a different-cluster reference patient (Adult#003 to Adult#008) (ablation) | model bootstrapping RL · Adult#008, Figure 5 | policy performance | No clear effect |
| switch to a closer same-cluster reference (Adult#002) for Adult#008 (ablation) | model bootstrapping RL · Adult#008, Figure 5 | policy performance | Improved |
| fine-tune transferred dynamics model on target patient data (ablation) | model bootstrapping RL · Simglucose policy learning | policy performance | Improved |
| direct policy transfer instead of training on target patient | ADRQN expert policy · Simglucose | long-term performance | Hurt |
Closest papers by meaning
- Hybrid$^2$ Neural ODE Causal Modeling and an Application to Glycemic Response (ICML 2024 · similarity 0.85)
- Safety Generalization Under Distribution Shift in Safe Reinforcement Learning: A Diabetes Testbed (ICML 2026 · similarity 0.84)
- GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting (NeurIPS 2026 · similarity 0.84)
- Data-Driven Discovery of Dynamical Systems in Pharmacology using Large Language Models (NeurIPS 2024 · similarity 0.81)
- Domain Adaptive Diabetic Retinopathy Grading with Model Absence and Flowing Data (CVPR 2025 · similarity 0.80)
- GlucoBench: Curated List of Continuous Glucose Monitoring Datasets with Prediction Benchmarks (ICLR 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
