Related Experiment Video
Updated: Oct 12, 2025

Oral Combinational Antiretroviral Treatment in HIV-1 Infected Humanized Mice
Published on: October 6, 2022
A reinforcement learning model to inform optimal decision paths for HIV elimination
Seyedeh N Khatami1, Chaitra Gopalappa1
1Mechanical and Industrial Engineering Department, University of Massachusetts Amherst, Amherst, MA 01003, USA.
Reinforcement learning models optimal HIV testing and retention-in-care rates to guide the Ending the HIV Epidemic initiative. While effective, these strategies alone may not meet 2030 targets, indicating a need for additional interventions.
Area of Science:
- Public Health
- Infectious Disease Modeling
- Artificial Intelligence in Healthcare
Background:
- The Ending the HIV Epidemic (EHE) initiative aims to significantly reduce HIV incidence in the US by 2030.
- Optimal combinations of HIV testing and retention-in-care are crucial for achieving EHE goals, but traditional optimization methods are infeasible due to disease complexity.
- Reinforcement learning (RL) offers a potential solution for complex public health decision-making, despite computational challenges.
Purpose of the Study:
- To evaluate the feasibility of using RL to identify optimal sequences of HIV testing and retention-in-care rates for HIV elimination.
- To determine if RL can overcome computational challenges in large-scale stochastic public health problems.
- To inform the implementation of EHE programs by providing data-driven strategies.
Main Methods:
- Trained an RL algorithm using a stochastic agent-based simulation to find optimal testing and retention-in-care rates over 5-year intervals (2015-2070).
- Defined optimality by maximizing quality-adjusted life-years and minimizing costs associated with HIV testing and treatment.
- Evaluated the sensitivity of optimal decisions to cost-function variations and addressed computational challenges through proxy decision-metrics.
Main Results:
- The RL model identified an optimal sequence, suggesting scaling up retention-in-care programs and initially high testing frequency, gradually decreasing over time as incidence declines.
- Results were robust to cost uncertainties, demonstrating the model's stability.
- The study confirmed RL's convergence and suitability for phased public health decision-making in infectious disease control.
Conclusions:
- HIV testing and retention-in-care strategies, optimized via RL, are essential but may be insufficient to meet the 2030 EHE targets alone.
- Additional interventions are likely necessary to achieve comprehensive HIV elimination.
- RL is a viable and powerful tool for optimizing complex, dynamic public health strategies for infectious disease control.
Related Concept Videos
Retrovirus Life Cycles
Decision Making: P-value Method
First, a specific claim about the population parameter is proposed. The claim is based on the research question and is stated in a simple form. Further, an opposing statement to the claim is also stated. These statements can act as null and alternative hypotheses: a null hypothesis would be a neutral statement while the alternative hypothesis can...
Parameters Affecting Nonlinear Elimination: Zero-Order Input, First-Order Absorption and Two-Compartment Model
When a drug is administered through a constant intravenous infusion and eliminated via nonlinear pharmacokinetics, it follows zero-order input. For example, oral drugs undergo first-order absorption upon administration and are eliminated through nonlinear pharmacokinetics.
In the case of subcutaneously administered drugs,...
Avoidance Learning and Learned Helplessness
Avoidance learning occurs when an organism learns that a specific behavior can prevent an unpleasant outcome. For example, a student who receives a bad grade may start studying harder to avoid future poor grades. This behavior persists even when the negative outcome is no longer present. Avoidance learning is powerful because it maintains behavior in the absence of the...
Decision Making
Automatic decision-making is fast, intuitive, and relies on gut feelings...
Decision Making: Traditional Method
First, a specific claim about the population parameter is decided based on the research question and is stated in a simple form. Further, an opposing statement to this claim is also stated. These statements can act as null and alternative hypotheses, out of which a null hypothesis would be a...

