Using machine learning to inform harvest control rule design in complex fishery settings

2024-12-16

arXiv.org

AI Summary

Managing the walleye fishery in Alberta, Canada faces challenges including high recruitment variability, complex age structure, and limited observational data. Method: This study proposes a data-driven harvest control rule (HCR) design framework that integrates proximal policy optimization (PPO)—a deep reinforcement learning algorithm—with Bayesian optimization, embedded within a stochastic age-structured population model. The framework jointly optimizes multi-objective management policies under partial observability and intermittent recruitment. A novel auxiliary observation—mean individual weight—is introduced to enhance policy robustness. Contribution/Results: The learned HCR significantly outperforms conventional biomass-based rectangular trigger policies across biological sustainability, economic yield, and management stability. It provides a generalizable methodological foundation for adaptive management of structurally complex fisheries.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

In fishery science, harvest management of size-structured stochastic populations is a long-standing and difficult problem. Rectilinear precautionary policies based on biomass and harvesting reference points have now become a standard approach to this problem. While these standard feedback policies are adapted from analytical or dynamic programming solutions assuming relatively simple ecological dynamics, they are often applied to more complicated ecological settings in the real world. In this paper we explore the problem of designing harvest control rules for partially observed, age-structured, spasmodic fish populations using tools from reinforcement learning (RL) and Bayesian optimization. Our focus is on the case of Walleye fisheries in Alberta, Canada, whose highly variable recruitment dynamics have perplexed managers and ecologists. We optimized and evaluated policies using several complementary performance metrics. The main questions we addressed were: 1. How do standard policies based on reference points perform relative to numerically optimized policies? 2. Can an observation of mean fish weight, in addition to stock biomass, aid policy decisions?

Problem

• Design harvest control rules for complex fish populations using reinforcement learning
• Evaluate performance of standard vs optimized policies in fishery management
• Assess impact of fish weight data on policy decisions

Innovation

• Using reinforcement learning for harvest control
• Applying Bayesian optimization in fishery management
• Optimizing policies with multiple performance metrics