Research map / Reinforcement learning / Visual and representation RL
Navigating Uncertainty in Epidemic Contexts with Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a simulation environment coupling stochastic epidemic models with reinforcement learning to learn classroom-occupancy policies, with a Q-learning proof of concept for the education domain.
Problem
Static epidemic response policies lack flexibility to adapt to changing outbreak conditions; there is a need for RL-derived policies that balance infection risk against in-person activity (e.g., classroom occupancy).
Approach
Integrates stochastic epidemic models (approximate SIR, indoor classroom model) as the state-transition dynamics of an RL environment, with a weighted reward trading off allowed students against expected infections, in an extensible simulation platform.
Key results
- Preliminary results suggest a Q-learning agent refines its policies through iterative training.
- Adjusting the reward weight parameter yields different Pareto-optimal trade-offs between minimizing student infections and maximizing on-campus presence.
Limitations
- Only a proof of concept with preliminary results; no quantitative evaluation reported.
- Future experiments are planned but not yet conducted.
Experimental setup
Metrics: expected infected students on-campus student presence
Builds on: Q-learning
Closest papers by meaning
- Optimizing Heat Alert Issuance with Reinforcement Learning (AAAI 2025 · similarity 0.79)
- Deterministic Uncertainty Propagation for Improved Model-Based Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.78)
- Stochastic Control for Fine-tuning Diffusion Models: Optimality, Regularity, and Convergence (ICML 2025 · similarity 0.78)
- Context in Public Health for Underserved Communities: A Bayesian Approach to Online Restless Bandits (AAAI 2025 · similarity 0.78)
- Sampling-Based Safe Reinforcement Learning (NeurIPS 2026 · similarity 0.78)
- Sequential Stochastic Combinatorial Optimization Using Hierarchal Reinforcement Learning (ICLR 2025 · similarity 0.78)
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.