Stratified Active Learning for Spatiotemporal Generalisation in Bioacoustic Monitoring

bioRxiv (Cold Spring Harbor Laboratory)

Abstract

Abstract Active learning optimises machine learning model training through the data-efficient selection of informative samples for annotation and training. In the context of biodiversity monitoring using passive acoustic monitoring, active learning offers a promising strategy to reduce the fundamental annotation bottleneck and improve global training efficiency. However, the generalisability of model performance across ecologically relevant strata (e.g. sites, season etc) is often overlooked. As passive acoustic monitoring is extended to larger scales and finer resolutions, inter-strata spatiotemporal variability also increases. We introduce and investigate the concept of stratified active learning to achieve reliable and generalisable model performance across deployment conditions. We compare between implicit cluster-based diversification methods and explicit stratification, demonstrating that cross-strata generalisation is a function of stratum divergence, not sampling balance. Additionally, mutual information as well as exclusion analysis show that spatiotemporal context can explain a substantial proportion of species label variance and inform sampling decisions.

Authors 3

1. Ben McEwen corresponding

   Tilburg University

   Affiliation as printed

   Tilburg University;

   Tilburg University, Netherlnds

2. Corentin Bernard

   Université de Toulon

   Affiliation as printed

   Toulon University

   Toulon University, France

3. Dan Stowell Aachen Naturalis Biodiversity Center

   Tilburg University · Naturalis Biodiversity Center

   Affiliation as printed

   Tilburg University;

   Naturalis Biodiversity Center, Netherlands

   Tilburg University, Netherlnds