An Online Learning-based Framework for Tracking

Kamalika Chaudhuri, Yoav Freund, Daniel Hsu
Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, PMLR R8:109-116, 2010.

Abstract

We study the tracking problem, namely, estimating the hidden state of an object over time, from unreliable and noisy measurements. The standard framework for the tracking problem is the generative framework, which is the basis of solutions such as the Bayesian algorithm and its approximation, the particle filters. However, these solutions can be very sensitive to model mismatches. In this paper, motivated by online learning, we introduce a new framework for tracking. We provide an efficient tracking algorithm for this framework. We provide experimental results comparing our algorithm to the Bayesian algorithm on simulated data. Our experiments show that when there are slight model mismatches, our algorithm outperforms the Bayesian algorithm.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR8-chaudhuri10a, title = {An Online Learning-based Framework for Tracking}, author = {Chaudhuri, Kamalika and Freund, Yoav and Hsu, Daniel}, booktitle = {Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence}, pages = {109--116}, 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/chaudhuri10a/chaudhuri10a.pdf}, url = {https://proceedings.mlr.press/r8/chaudhuri10a.html}, abstract = {We study the tracking problem, namely, estimating the hidden state of an object over time, from unreliable and noisy measurements. The standard framework for the tracking problem is the generative framework, which is the basis of solutions such as the Bayesian algorithm and its approximation, the particle filters. However, these solutions can be very sensitive to model mismatches. In this paper, motivated by online learning, we introduce a new framework for tracking. We provide an efficient tracking algorithm for this framework. We provide experimental results comparing our algorithm to the Bayesian algorithm on simulated data. Our experiments show that when there are slight model mismatches, our algorithm outperforms the Bayesian algorithm.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T An Online Learning-based Framework for Tracking %A Kamalika Chaudhuri %A Yoav Freund %A Daniel Hsu %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-chaudhuri10a %I PMLR %P 109--116 %U https://proceedings.mlr.press/r8/chaudhuri10a.html %V R8 %X We study the tracking problem, namely, estimating the hidden state of an object over time, from unreliable and noisy measurements. The standard framework for the tracking problem is the generative framework, which is the basis of solutions such as the Bayesian algorithm and its approximation, the particle filters. However, these solutions can be very sensitive to model mismatches. In this paper, motivated by online learning, we introduce a new framework for tracking. We provide an efficient tracking algorithm for this framework. We provide experimental results comparing our algorithm to the Bayesian algorithm on simulated data. Our experiments show that when there are slight model mismatches, our algorithm outperforms the Bayesian algorithm. %Z Reissued by PMLR on 04 October 2026.
APA
Chaudhuri, K., Freund, Y. & Hsu, D.. (2010). An Online Learning-based Framework for Tracking. Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R8:109-116 Available from https://proceedings.mlr.press/r8/chaudhuri10a.html. Reissued by PMLR on 04 October 2026.

Related Material