Regularized State Estimation And Parameter Learning Via Augmented Lagrangian Kalman Smoother Method
(2019) IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)- Abstract
- In this article, we address the problem of estimating the state and learning of the parameters in a linear dynamic system with generalized L 1 -regularization. Assuming a sparsity prior on the state, the joint state estimation and parameter learning problem is cast as an unconstrained optimization problem. However, when the dimensionality of state or parameters is large, memory requirements and computation of learning algorithms are generally prohibitive. Here, we develop a new augmented Lagrangian Kalman smoother method for solving this problem, where the primal variable update is reformulated as Kalman smoother. The effectiveness of the proposed method for state estimation and parameter learning is demonstrated in spectro-temporal... (More)
- In this article, we address the problem of estimating the state and learning of the parameters in a linear dynamic system with generalized L 1 -regularization. Assuming a sparsity prior on the state, the joint state estimation and parameter learning problem is cast as an unconstrained optimization problem. However, when the dimensionality of state or parameters is large, memory requirements and computation of learning algorithms are generally prohibitive. Here, we develop a new augmented Lagrangian Kalman smoother method for solving this problem, where the primal variable update is reformulated as Kalman smoother. The effectiveness of the proposed method for state estimation and parameter learning is demonstrated in spectro-temporal estimation tasks using both synthetic and real data. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/61cd841c-1d1d-42c9-b5b5-964f82628c3e
- author
- Gao, Rui ; Tronarp, Filip LU ; Zhao, Zheng and Särkkä, Simo
- publishing date
- 2019
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)
- conference location
- Pittsburgh, United States
- conference dates
- 2019-10-13 - 2019-10-16
- external identifiers
-
- scopus:85077706727
- DOI
- 10.1109/MLSP.2019.8918821
- language
- English
- LU publication?
- no
- id
- 61cd841c-1d1d-42c9-b5b5-964f82628c3e
- date added to LUP
- 2023-08-20 23:00:22
- date last changed
- 2023-11-10 14:09:29
@inproceedings{61cd841c-1d1d-42c9-b5b5-964f82628c3e, abstract = {{In this article, we address the problem of estimating the state and learning of the parameters in a linear dynamic system with generalized L 1 -regularization. Assuming a sparsity prior on the state, the joint state estimation and parameter learning problem is cast as an unconstrained optimization problem. However, when the dimensionality of state or parameters is large, memory requirements and computation of learning algorithms are generally prohibitive. Here, we develop a new augmented Lagrangian Kalman smoother method for solving this problem, where the primal variable update is reformulated as Kalman smoother. The effectiveness of the proposed method for state estimation and parameter learning is demonstrated in spectro-temporal estimation tasks using both synthetic and real data.}}, author = {{Gao, Rui and Tronarp, Filip and Zhao, Zheng and Särkkä, Simo}}, booktitle = {{IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)}}, language = {{eng}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Regularized State Estimation And Parameter Learning Via Augmented Lagrangian Kalman Smoother Method}}, url = {{http://dx.doi.org/10.1109/MLSP.2019.8918821}}, doi = {{10.1109/MLSP.2019.8918821}}, year = {{2019}}, }