Reinforcement Learning Interventions on Boundedly Rational Human Agents in Frictionful Tasks
Not on the map. Why?
Abstract (arXiv)
Many important behavior changes are frictionful; they require individuals to expend effort over a long period with little immediate gratification. Here, an artificial intelligence (AI) agent can provide personalized interventions to help individuals stick to their goals. In these settings, the AI agent must personalize rapidly (before the individual disengages) and interpretably, to help us understand the behavioral interventions. In this paper, we introduce Behavior Model Reinforcement Learning (BMRL), a framework in which an AI agent intervenes on the parameters of a Markov Decision Process (MDP) belonging to a boundedly rational human agent. Our formulation of the human decision-maker as a planning agent allows us to attribute undesirable human policies (ones that do not lead to the goal) to their maladapted MDP parameters, such as an extremely low discount factor. Furthermore, we propose a class of tractable human models that captures fundamental behaviors in frictionful tasks. Introducing a notion of MDP equivalence specific to BMRL, we theoretically and empirically show that AI planning with our human models can lead to helpful policies on a wide range of more complex, ground-truth humans.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 13 of its 50 references that are in the map. First posted on arXiv 26 Jan 2024.
Builds on
13 of 50 references.
References · 13
- 2019Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observationscited by 256
- 2020Online Bayesian Goal Inference for Boundedly-Rational Planning Agentscited by 67
- 2020Plannable Approximations to MDP Homomorphisms: Equivariance under Actionscited by 61
- 2019Data Efficient Reinforcement Learning for Legged Robotscited by 60
- 2019Hyperbolic Discounting and Learning over Multiple Horizonscited by 58
- 2019On the Feasibility of Learning, Rather than Assuming, Human Biases for Reward Inferencecited by 49
- 2022Designing Reinforcement Learning Algorithms for Digital Interventions: Pre-implementation Guidelinescited by 22
- 2020Assisted Perception: Optimizing Observations to Communicate Statecited by 10
- 2020Sample-efficient Reinforcement Learning in Robotic Table Tenniscited by 10
- 2023Inverse Decision Modeling: Learning Interpretable Representations of Behaviorcited by 10
- 2019Sample-efficient Deep Reinforcement Learning with Imaginary Rollouts for Human-Robot Interactioncited by 3
- 2019Modeling and Interpreting Real-world Human Risk Decision Making with Inverse Reinforcement Learningcited by 2
Built on by
Citing papers per year: 2024: 1; 2025: 0; 2026: 1.
2 papers.