A Maximum Entropy Approach to Species Distribution Modeling
Abstract
We study the problem of modeling species geographic distributions, a criticalproblem in conservation biology. We propose the use of maximum-entropytechniques for this problem, specifically, sequential-update algorithms thatcan handle a very large number of features. We describe experiments comparingmaxent with a standard distribution-modeling tool, called GARP, on a dataset containing observationdata for North American breeding birds. We also study how well maxent performsas a function of the number of training examples and training time, analyzethe use of regularization to avoid overfitting when the number of examples issmall, and explore the interpretability of models constructed using maxent.
Cite
Text
Phillips et al. "A Maximum Entropy Approach to Species Distribution Modeling." International Conference on Machine Learning, 2004. doi:10.1145/1015330.1015412Markdown
[Phillips et al. "A Maximum Entropy Approach to Species Distribution Modeling." International Conference on Machine Learning, 2004.](https://mlanthology.org/icml/2004/phillips2004icml-maximum/) doi:10.1145/1015330.1015412BibTeX
@inproceedings{phillips2004icml-maximum,
title = {{A Maximum Entropy Approach to Species Distribution Modeling}},
author = {Phillips, Steven J. and Dudík, Miroslav and Schapire, Robert E.},
booktitle = {International Conference on Machine Learning},
year = {2004},
doi = {10.1145/1015330.1015412},
url = {https://mlanthology.org/icml/2004/phillips2004icml-maximum/}
}