Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session B: Wednesday, August 13, 1:00 – 4:00 pm, de Brug & E‑Hall

Optimal Foraging by Learning the World Model

Roxana Zeraati¹, Tiffany Oña Jodar, Shervin Safavi², Bruno Cruz, Cindy Poo³, Peter Dayan⁴; ¹Max-Planck Institute for Biological Cybernetics, ²Technische Universität Dresden, ³Allen Institute, ⁴Max-Planck Institute

Presenter: Roxana Zeraati

Patch foraging—deciding when to leave a depleting resource to search for alternatives—is a fundamental aspect of animal behavior and offers a window into ethologically grounded decision processes. Several theories, most notably the Marginal Value Theorem (MVT), have proposed strategies for optimal foraging. However, they typically ignore most details of the spatiotemporal structure of the environment, and particularly the dynamics of the replenishment of patches. We investigate optimal patch foraging with richer replenishment timescales. Using average-reward reinforcement learning (RL), we show that under slow replenishment, optimal policies leverage the world model to generate higher reward rates and distinct behavioral statistics from MVT and similar policies. Our results provide testable predictions for future experiments.

Topic Area: Reward, Value & Social Decision Making

Extended Abstract: Full Text PDF