Publications

Active Learning-Based Species Range Estimation

Published in Advances in Neural Information Processing Systems 36, 2023

Here we investigate a new active learning based approach for the prediction of a species’ range from limited data and evaluate our approach on two ground truth datasets. Our approach makes use of species distribution models for well known species in order to improve performance for our target species.

Code here

Download here

Spatial Implicit Neural Representations for Global-Scale Species Mapping

Published in Fortieth International Conference on Machine Learning, 2023

Here we investigate implicit neural representations for jointly estimating the spatial range of thousands of species from noisy, sparse, community collected data, with new benchmark tasks for geospatial representations and species distribution models.

Demo here

Website here

Download here