Bayesian inverse reinforcement learning for collective animal movement
Links
- More information: Publisher Index Page (via DOI)
- Open Access Version: External Repository
- Download citation as: RIS | Dublin Core
Abstract
Agent-based methods allow for defining simple rules that generate complex group behaviors. The governing rules of such models are typically set a priori, and parameters are tuned from observed behavior trajectories. Instead of making simplifying assumptions across all anticipated scenarios, inverse reinforcement learning provides inference on the short-term (local) rules governing long-term behavior policies by using properties of a Markov decision process. We use the computationally efficient linearly-solvable Markov decision process to learn the local rules governing collective movement for a simulation of the selfpropelled-particle (SPP) model and a data application for a captive guppy population. The estimation of the behavioral decision costs is done in a Bayesian framework with basis function smoothing. We recover the true costs in the SPP simulation and find the guppies value collective movement more than targeted movement toward shelter.
Suggested Citation
Schafer, T.L., Wikle, C., Hooten, M., 2022, Bayesian inverse reinforcement learning for collective animal movement: Annals of Applied Statistics, v. 16, no. 2, p. 999-1013, https://doi.org/10.1214/21-AOAS1529.
| Publication type | Article |
|---|---|
| Publication Subtype | Journal Article |
| Title | Bayesian inverse reinforcement learning for collective animal movement |
| Series title | Annals of Applied Statistics |
| DOI | 10.1214/21-AOAS1529 |
| Volume | 16 |
| Issue | 2 |
| Year Published | 2022 |
| Language | English |
| Publisher | Project Euclid |
| Contributing office(s) | Coop Res Unit Seattle |
| Description | 15 p. |
| First page | 999 |
| Last page | 1013 |