About this code

This is the backend code for the RL algorithm used in the Oralytics clinical trial. The RL algorithm in Oralytics optimizes the delivery of engagement prompts to participants to maximize their brushing quality (or oral self-care behaviors). For more information about Oralytics, please see the linked manuscripts.

How can a behavioral scientist use this code?

Behavioral intervention scientists can use this code as a starting point for optimizing the delivery of mobile-based prompts to encourage oral hygiene behaviors.

Related References

Trella, A. L., Zhang, K. W., Nahum-Shani, I., Shetty, V., Doshi-Velez, F., & Murphy, S. A. (2022). Designing reinforcement learning algorithms for digital interventions: Pre-implementation guidelines. Algorithms, 15(8), 255. doi:https://doi.org/10.3390/a15080255

Trella, A. L., Zhang, K. W., Nahum-Shani, I., Shetty, V., Doshi-Velez, F., & Murphy, S. A. (2023). Reward Design for an Online Reinforcement Learning Algorithm Supporting Oral Self-Care. Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 15724-15730. https://doi.org/10.1609/aaai.v37i13.26866