Combining Spatial and Telemetric Features for Learning Animal Movement Models

Berk Kapicioglu, Robert Schapire, Martin Wikelski, Tamara Broderick
Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, PMLR R8:268-275, 2010.

Abstract

We introduce a new graphical model for tracking radio-tagged animals and learning their movement patterns. The model pro- vides a principled way to combine radio telemetry data with an arbitrary set of user- defined, spatial features. We describe an ef- ficient stochastic gradient algorithm for fit- ting model parameters to data and demon- strate its effectiveness via asymptotic analy- sis and synthetic experiments. We also ap- ply our model to real datasets, and show that it outperforms the most popular ra- dio telemetry software package used in ecol- ogy. We conclude that integration of dif- ferent data sources under a single statistical framework, coupled with appropriate param- eter and state estimation procedures, pro- duces both accurate location estimates and an interpretable statistical model of animal movement.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR8-kapicioglu10a, title = {Combining Spatial and Telemetric Features for Learning Animal Movement Models}, author = {Kapicioglu, Berk and Schapire, Robert and Wikelski, Martin and Broderick, Tamara}, booktitle = {Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence}, pages = {268--275}, year = {2010}, editor = {Grünwald, Peter and Spirtes, Peter}, volume = {R8}, series = {Proceedings of Machine Learning Research}, month = {08--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r8/main/assets/kapicioglu10a/kapicioglu10a.pdf}, url = {https://proceedings.mlr.press/r8/kapicioglu10a.html}, abstract = {We introduce a new graphical model for tracking radio-tagged animals and learning their movement patterns. The model pro- vides a principled way to combine radio telemetry data with an arbitrary set of user- defined, spatial features. We describe an ef- ficient stochastic gradient algorithm for fit- ting model parameters to data and demon- strate its effectiveness via asymptotic analy- sis and synthetic experiments. We also ap- ply our model to real datasets, and show that it outperforms the most popular ra- dio telemetry software package used in ecol- ogy. We conclude that integration of dif- ferent data sources under a single statistical framework, coupled with appropriate param- eter and state estimation procedures, pro- duces both accurate location estimates and an interpretable statistical model of animal movement.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Combining Spatial and Telemetric Features for Learning Animal Movement Models %A Berk Kapicioglu %A Robert Schapire %A Martin Wikelski %A Tamara Broderick %B Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2010 %E Peter Grünwald %E Peter Spirtes %F pmlr-vR8-kapicioglu10a %I PMLR %P 268--275 %U https://proceedings.mlr.press/r8/kapicioglu10a.html %V R8 %X We introduce a new graphical model for tracking radio-tagged animals and learning their movement patterns. The model pro- vides a principled way to combine radio telemetry data with an arbitrary set of user- defined, spatial features. We describe an ef- ficient stochastic gradient algorithm for fit- ting model parameters to data and demon- strate its effectiveness via asymptotic analy- sis and synthetic experiments. We also ap- ply our model to real datasets, and show that it outperforms the most popular ra- dio telemetry software package used in ecol- ogy. We conclude that integration of dif- ferent data sources under a single statistical framework, coupled with appropriate param- eter and state estimation procedures, pro- duces both accurate location estimates and an interpretable statistical model of animal movement. %Z Reissued by PMLR on 04 October 2026.
APA
Kapicioglu, B., Schapire, R., Wikelski, M. & Broderick, T.. (2010). Combining Spatial and Telemetric Features for Learning Animal Movement Models. Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R8:268-275 Available from https://proceedings.mlr.press/r8/kapicioglu10a.html. Reissued by PMLR on 04 October 2026.

Related Material