Robust Online Model Adaptation by Extended Kalman Filter with Exponential Moving Average and Dynamic Multi-Epoch Strategy

Abulikemu Abuduweili, Changliu Liu
Proceedings of the 2nd Conference on Learning for Dynamics and Control, PMLR 120:65-74, 2020.

Abstract

High fidelity behavior prediction of intelligent agents is critical in many applications. However, the prediction model trained on the training set may not generalize to the testing set due to domain shift and time variance. The challenge motivates the adoption of online adaptation algorithms to update prediction models in real-time to improve the prediction performance. Inspired by Extended Kalman Filter (EKF), this paper introduces a series of online adaptation methods, which are applicable to neural network-based models. A base adaptation algorithm Modified EKF with forgetting factor (MEKF_lambda) is introduced first, followed by exponential moving average filtering techniques. Then this paper introduces a dynamic multi-epoch update strategy to effectively utilize samples received in real time. With all these extensions, we propose a robust online adaptation algorithm: MEKF with Exponential Moving Average and Dynamic Multi-Epoch strategy (MEKF_EMA-DME). The proposed algorithm outperforms existing methods as demonstrated in experiments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v120-abuduweili20a, title = {Robust Online Model Adaptation by Extended Kalman Filter with Exponential Moving Average and Dynamic Multi-Epoch Strategy}, author = {Abuduweili, Abulikemu and Liu, Changliu}, booktitle = {Proceedings of the 2nd Conference on Learning for Dynamics and Control}, pages = {65--74}, year = {2020}, editor = {Bayen, Alexandre M. and Jadbabaie, Ali and Pappas, George and Parrilo, Pablo A. and Recht, Benjamin and Tomlin, Claire and Zeilinger, Melanie}, volume = {120}, series = {Proceedings of Machine Learning Research}, month = {10--11 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v120/abuduweili20a/abuduweili20a.pdf}, url = {https://proceedings.mlr.press/v120/abuduweili20a.html}, abstract = {High fidelity behavior prediction of intelligent agents is critical in many applications. However, the prediction model trained on the training set may not generalize to the testing set due to domain shift and time variance. The challenge motivates the adoption of online adaptation algorithms to update prediction models in real-time to improve the prediction performance. Inspired by Extended Kalman Filter (EKF), this paper introduces a series of online adaptation methods, which are applicable to neural network-based models. A base adaptation algorithm Modified EKF with forgetting factor (MEKF_lambda) is introduced first, followed by exponential moving average filtering techniques. Then this paper introduces a dynamic multi-epoch update strategy to effectively utilize samples received in real time. With all these extensions, we propose a robust online adaptation algorithm: MEKF with Exponential Moving Average and Dynamic Multi-Epoch strategy (MEKF_EMA-DME). The proposed algorithm outperforms existing methods as demonstrated in experiments.} }
Endnote
%0 Conference Paper %T Robust Online Model Adaptation by Extended Kalman Filter with Exponential Moving Average and Dynamic Multi-Epoch Strategy %A Abulikemu Abuduweili %A Changliu Liu %B Proceedings of the 2nd Conference on Learning for Dynamics and Control %C Proceedings of Machine Learning Research %D 2020 %E Alexandre M. Bayen %E Ali Jadbabaie %E George Pappas %E Pablo A. Parrilo %E Benjamin Recht %E Claire Tomlin %E Melanie Zeilinger %F pmlr-v120-abuduweili20a %I PMLR %P 65--74 %U https://proceedings.mlr.press/v120/abuduweili20a.html %V 120 %X High fidelity behavior prediction of intelligent agents is critical in many applications. However, the prediction model trained on the training set may not generalize to the testing set due to domain shift and time variance. The challenge motivates the adoption of online adaptation algorithms to update prediction models in real-time to improve the prediction performance. Inspired by Extended Kalman Filter (EKF), this paper introduces a series of online adaptation methods, which are applicable to neural network-based models. A base adaptation algorithm Modified EKF with forgetting factor (MEKF_lambda) is introduced first, followed by exponential moving average filtering techniques. Then this paper introduces a dynamic multi-epoch update strategy to effectively utilize samples received in real time. With all these extensions, we propose a robust online adaptation algorithm: MEKF with Exponential Moving Average and Dynamic Multi-Epoch strategy (MEKF_EMA-DME). The proposed algorithm outperforms existing methods as demonstrated in experiments.
APA
Abuduweili, A. & Liu, C.. (2020). Robust Online Model Adaptation by Extended Kalman Filter with Exponential Moving Average and Dynamic Multi-Epoch Strategy. Proceedings of the 2nd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 120:65-74 Available from https://proceedings.mlr.press/v120/abuduweili20a.html.

Related Material