Smart AI coach helps smokers quit using Reinforcement Learning

Nele Albers of Delft University of Technology (The Netherlands) uses AI coaches to help smokers quit smoking and vaping for her doctoral dissertation. She developed an AI coach that uses psychologically informed Reinforcement Learning (RL), a form of machine learning in which a model learns through rewards, similar to how humans learn behaviors. Albers based her research on insights from behavior change theories and data from three large-scale studies with more than 500 participants each. 

In addition to the technical side, Albers examined ethical, economic and psychological aspects and analyzed how different factors contribute to effective behavior change, namely: how to convince smokers to quit, what they are asked to do, and who should support them. Her research shows that AI coaches who take psychological principles into account have great potential to provide effective support for people who want to quit smoking.

“My research shows how personalizing support - by taking into account both a person's current and future state - increases the effectiveness of AI-based eHealth applications. This offers many opportunities for behavioral change.”

Smarter eHealth applications through personalized AI coaching
AI coaches in eHealth applications could effectively guide people through behavior change while also cutting costs. Yet these applications are not yet widely deployed. Challenges are, for example, dropout and lack of engagement. Better-tailored support -- which takes knowledge, motivation, vitality, thinking patterns into account -- can systematically increase its usage. To improve the effectiveness of support to quit smoking, Albers examined both the algorithmic side and the interaction between smokers and AI coaches. For example, her model can determine when it is useful to involve a human coach in addition to the AI coach. She also looked at the tension between smokers' preferences and the advice of health experts. Health experts determine what is right, but smokers often look at this differently. Albers' algorithm tries to find a balance between the two perspectives.

Research findings
A key finding from her research is that support is more effective when it is adapted to an individual's situation, such as when suggesting different activities. For example, the AI coach can encourage someone to think about stimuli that trigger the desire to smoke or the person they want to be in the future. Consider a motivation such as, “I want to be a better parent to my child by setting a good example.” In addition, context plays a role in the effectiveness of messages. Depending on the situation, wording such as “Other smokers find...” works better than “Doctors recommend.” Based on her analysis and findings, she developed models that increase smoker engagement and provide better support in building skills to quit.

Press inquiries and more information
Upon request, it is possible for the press to experience the research of Nele Albers up close by attending a session with the AI coach themselves. This can be done in the context of preparing to quit smoking, where the AI coach offers support. To get access contact Marc de Kool, Digital Society press officer, a.j.m.dekool@tudelft.nl / 06-10038065.