Search papers, labs, and topics across Lattice.
This paper introduces LivestockFL, a federated learning framework for livestock growth prediction, enabling collaborative model training across farms without direct data sharing. The framework uses a GRU combined with an MLP to model temporal growth patterns. Results on a real-world dataset show that LivestockFL and its personalized variant, LivestockPFL, effectively predict livestock growth while preserving data privacy.
Federated learning can overcome data sparsity and privacy concerns to improve livestock growth prediction using real-world farm data.
Livestock growth prediction is essential for optimising farm management and improving the efficiency and sustainability of livestock production, yet it remains underexplored due to limited large-scale datasets and privacy concerns surrounding farm-level data. Existing biophysical models rely on fixed formulations, while most machine learning approaches are trained on small, isolated datasets, limiting their robustness and generalisability. To address these challenges, we propose LivestockFL, the first federated learning framework specifically designed for livestock growth prediction. LivestockFL enables collaborative model training across distributed farms without sharing raw data, thereby preserving data privacy while alleviating data sparsity, particularly for farms with limited historical records. The framework employs a neural architecture based on a Gated Recurrent Unit combined with a multilayer perceptron to model temporal growth patterns from historical weight records and auxiliary features. We further introduce LivestockPFL, a novel personalised federated learning framework that extends the above federated learning framework with a personalized prediction head trained on each farm's local data, producing farm-specific predictors. Experiments on a real-world dataset demonstrate the effectiveness and practicality of the proposed approaches.
