Online Program

Return to main conference page

All Times EDT

, -
Virtual
Contributed Presentations

Inverse Reinforcement Learning for Animal Movement Data (309920)

Mevin Hooten, Colorado State University 
*Toryn Schafer, Cornell University 
Christopher Wikle, University of Missouri 

Keywords: reinforcement learning, Bayesian learning, animal movement, Markov decision process

Inverse reinforcement learning provides inference on the short-term (local) rules governing long term behavior policies by using properties of a Markov decision process. Animal movement models are inferential tools for data collected by telemetry devices on animals. We demonstrate the Markov decision process as an alternative modeling framework for animal movement data which allows for memory and complex interactions between the environment and individuals. Model fitting is done in a Bayesian framework with approximate inference obtained by variational inference in large state space settings. Lastly, we compare inference to another common animal movement model.