Active Learning-Based Species Range Estimation
This work addresses species range estimation for ecologists and conservationists, offering an incremental improvement by integrating active learning with transfer learning from large datasets.
The paper tackles the problem of efficiently estimating a species' geographic range from limited observations by proposing an active learning approach that uses transfer-learned spatial representations from crowdsourced data, and demonstrates that it outperforms existing active learning methods and approaches end-to-end model performance with less data.
We propose a new active learning approach for efficiently estimating the geographic range of a species from a limited number of on the ground observations. We model the range of an unmapped species of interest as the weighted combination of estimated ranges obtained from a set of different species. We show that it is possible to generate this candidate set of ranges by using models that have been trained on large weakly supervised community collected observation data. From this, we develop a new active querying approach that sequentially selects geographic locations to visit that best reduce our uncertainty over an unmapped species' range. We conduct a detailed evaluation of our approach and compare it to existing active learning methods using an evaluation dataset containing expert-derived ranges for one thousand species. Our results demonstrate that our method outperforms alternative active learning methods and approaches the performance of end-to-end trained models, even when only using a fraction of the data. This highlights the utility of active learning via transfer learned spatial representations for species range estimation. It also emphasizes the value of leveraging emerging large-scale crowdsourced datasets, not only for modeling a species' range, but also for actively discovering them.