Scientific Findings

Get insights from Lirio behavioral science and AI experts on how to move people toward healthy behaviors and actions that drive positive outcomes.

BeSci Briefs

Our BeSci briefs cover the behavioral science solutions Lirio uses to optimize the content, channel, and timing of behavioral interventions and the behavioral biases that inform them.

Curiosity – Lirio Bias Brief

Curiosity – Lirio Bias Brief

Lirio delivers results by drawing insights from hundreds of behavioral biases and theories. Our Bias Brief series walks through specific biases one by one, each thoughtfully selected from our long list of insights. Read on for a brief rundown on the chosen bias,...

Educational Sessions

Our team has led informative discussions and presentations about the power of combining behavioral science and AI to create behavior change.

Solving From the Outside In: the Art of UX Design

Solving From the Outside In: the Art of UX Design

The Behavior Change Podcast by Lirio explores the various ways humans can leverage behavioral science to personalize our messaging, engage our audience, and drive better behavior at scale. Host: Greg Stielstra (GS), Senior Director of Behavioral Science at Lirio...

Academic Research

Our team has developed and presented a range of academic research pieces across the fields of behavioral science and AI.

Artificial Intelligence

Reinforcement Learning

Lirio Research: A Novel Policy Comparison Metric for Reinforcement Learning

Lirio’s Behavioral Reinforcement Learning Lab (BReLL) recently published a paper describing a new approach, the Limited Data Estimator, for comparing reinforcement learning policies using limited historical data.

Reinforcement learning

Lirio Research: Introducing A New Form of Machine Learning Optimization

Lirio’s AI Research team recently developed a novel adaptive stochastic gradient-free (ASGF) approach for solving some of the most difficult optimization challenges in machine learning. This innovative optimization algorithm, which is simple to implement and does not require careful fine-tuning, offers significant improvements when compared to existing state-of-the-art approaches.