Summary of Modeling and Optimization Of Epidemiological Control Policies Through Reinforcement Learning, by Ishir Rao
Modeling and Optimization of Epidemiological Control Policies Through Reinforcement Learning
by Ishir Rao
First submitted to arxiv on: 25 Jan 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Populations and Evolution (q-bio.PE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research proposes a multi-objective reinforcement learning (MORL) model to design optimal pandemic control strategies that balance the reduction of infection rates with economic impact. By combining an epidemiological Susceptible, Exposed, Infected, Recovered, Deceased (SEIRD) model with a deep double recurrent Q-network, the authors trained a reinforcement learning agent to enforce restrictions on the SEIRD simulation based on a reward function. Two agents were tested, each with unique reward functions and pandemic goals, resulting in two strategies: one prioritizing long lockdowns followed by cyclical shorter lockdowns, and another implementing 10-day lockdowns and 20-day no-restriction cycles. This approach allows for both economic and infection mitigation in multiple pandemic scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Pandemics are big problems that affect many people around the world. They can cause a lot of sickness and death, but also damage to economies. To help minimize these effects, experts use special models called epidemiological models. These models suggest ways to control pandemics using things like social distancing, curfews, and lockdowns. However, designing these strategies is tricky because it involves balancing the need to reduce infection rates with the need to protect economies. This research used a type of AI called multi-objective reinforcement learning (MORL) to design optimal pandemic control strategies. The authors combined an epidemiological model with a special AI agent that learned how to enforce restrictions based on a reward function. They tested two agents, each with different goals and priorities, resulting in two different strategies for controlling pandemics. |
Keywords
» Artificial intelligence » Reinforcement learning
