Sensor Selection for Crowdsensing Dynamical Systems

Francois Schnitzler, Jia Yuan Yu, Shie Mannor
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:829-837, 2015.

Abstract

We model crowdsensing as the selection of sensors with unknown variance to monitor a large linear dynamical system. To achieve low estimation error, we propose a Thompson sampling approach combining submodular optimization and a scalable online variational inference algorithm to maintain the posterior distribution over the variance. We also consider three alternative parameter estimation algorithms. We illustrate the behavior of our sensor selection algorithms on real traffic data from the city of Dublin. Our online algorithm achieves significantly lower estimation error than sensor selection using a fixed variance value for all sensors.

Cite this Paper

BibTeX
@InProceedings{pmlr-v38-schnitzler15, title = {{Sensor Selection for Crowdsensing Dynamical Systems}}, author = {Schnitzler, Francois and Yuan Yu, Jia and Mannor, Shie}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {829--837}, year = {2015}, editor = {Lebanon, Guy and Vishwanathan, S. V. N.}, volume = {38}, series = {Proceedings of Machine Learning Research}, address = {San Diego, California, USA}, month = {09--12 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v38/schnitzler15.pdf}, url = {https://proceedings.mlr.press/v38/schnitzler15.html}, abstract = {We model crowdsensing as the selection of sensors with unknown variance to monitor a large linear dynamical system. To achieve low estimation error, we propose a Thompson sampling approach combining submodular optimization and a scalable online variational inference algorithm to maintain the posterior distribution over the variance. We also consider three alternative parameter estimation algorithms. We illustrate the behavior of our sensor selection algorithms on real traffic data from the city of Dublin. Our online algorithm achieves significantly lower estimation error than sensor selection using a fixed variance value for all sensors.} }
Endnote
%0 Conference Paper %T Sensor Selection for Crowdsensing Dynamical Systems %A Francois Schnitzler %A Jia Yuan Yu %A Shie Mannor %B Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2015 %E Guy Lebanon %E S. V. N. Vishwanathan %F pmlr-v38-schnitzler15 %I PMLR %P 829--837 %U https://proceedings.mlr.press/v38/schnitzler15.html %V 38 %X We model crowdsensing as the selection of sensors with unknown variance to monitor a large linear dynamical system. To achieve low estimation error, we propose a Thompson sampling approach combining submodular optimization and a scalable online variational inference algorithm to maintain the posterior distribution over the variance. We also consider three alternative parameter estimation algorithms. We illustrate the behavior of our sensor selection algorithms on real traffic data from the city of Dublin. Our online algorithm achieves significantly lower estimation error than sensor selection using a fixed variance value for all sensors.
RIS
TY - CPAPER TI - Sensor Selection for Crowdsensing Dynamical Systems AU - Francois Schnitzler AU - Jia Yuan Yu AU - Shie Mannor BT - Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics DA - 2015/02/21 ED - Guy Lebanon ED - S. V. N. Vishwanathan ID - pmlr-v38-schnitzler15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 38 SP - 829 EP - 837 L1 - http://proceedings.mlr.press/v38/schnitzler15.pdf UR - https://proceedings.mlr.press/v38/schnitzler15.html AB - We model crowdsensing as the selection of sensors with unknown variance to monitor a large linear dynamical system. To achieve low estimation error, we propose a Thompson sampling approach combining submodular optimization and a scalable online variational inference algorithm to maintain the posterior distribution over the variance. We also consider three alternative parameter estimation algorithms. We illustrate the behavior of our sensor selection algorithms on real traffic data from the city of Dublin. Our online algorithm achieves significantly lower estimation error than sensor selection using a fixed variance value for all sensors. ER -
APA
Schnitzler, F., Yuan Yu, J. & Mannor, S.. (2015). Sensor Selection for Crowdsensing Dynamical Systems. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 38:829-837 Available from https://proceedings.mlr.press/v38/schnitzler15.html.

Related Material