arXiv subjects

Abigail G. Keller

Publications and source records attributed to Abigail G. Keller.

2 recordsLinked to original sources

arXiv2026

Bayesian decision theory for wildlife management under uncertainty: from inference to action

Ecologists are increasingly expected to inform management decisions under uncertainty, yet most analytical workflows stop at statistical inference. Bayesian decision theory provides a coherent framework to bridge this gap by propagating posterior uncertainty to evaluate alternative actions through utility functions, but remains underused in ecology. Here, we present a practical workflow for implementing Bayesian decision theory using standard Bayesian tools, illustrated with two case studies: wolf management in France, where the decision is the number of wolves to remove under uncertain population dynamics, and invasive muskrat management in the Netherlands, where control effort is allocated across space. In both cases, expected utility integrates posterior uncertainty and management trade-offs. Optimal decisions emerge as compromises between competing objectives. For wolves, optimal harvest balances removal benefits and population risk. For muskrats, optimal effort increases with the importance of population reduction and is unevenly allocated across provinces. Bayesian decision theory provides a formal interface between scientists, who characterize ecological systems and uncertainty, and decision-makers, who define objectives, values and trade-offs. By making these trade-offs explicit, it enhances transparency and relevance for management. It also provides a common framework for bringing together Bayesian statistics, decision analysis and risk analysis, strengthening the link between ecological inference and action.

stat.AP↗

arXiv2023

Pretty darn good control: when are approximate solutions better than approximate models

Existing methods for optimal control struggle to deal with the complexity commonly encountered in real-world systems, including dimensionality, process error, model bias and data heterogeneity. Instead of tackling these system complexities directly, researchers have typically sought to simplify models to fit optimal control methods. But when is the optimal solution to an approximate, stylized model better than an approximate solution to a more accurate model? While this question has largely gone unanswered owing to the difficulty of finding even approximate solutions for complex models, recent algorithmic and computational advances in deep reinforcement learning (DRL) might finally allow us to address these questions. DRL methods have to date been applied primarily in the context of games or robotic mechanics, which operate under precisely known rules. Here, we demonstrate the ability for DRL algorithms using deep neural networks to successfully approximate solutions (the "policy function" or control rule) in a non-linear three-variable model for a fishery without knowing or ever attempting to infer a model for the process itself. We find that the reinforcement learning agent discovers an effective simplification of the problem to obtain an interpretable control rule. We show that the policy obtained with DRL is both more profitable and more sustainable than any constant mortality policy -- the standard family of policies considered in fishery management.

cs.LG↗