Bayesian Continual Imputation and Prediction For Irregularly Sampled Time Series Data. Guo, Y., Jun Poh, J. W., Wong, C. S. Y., & Ramasamy, S. In ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 4493–4497, May, 2022.
doi  abstract   bibtex   
Learning from irregularly sampled, streaming, multi-variate time-series data with many missing values is a very challenging task. In this paper, we propose a Bayesian Continual Imputation and Prediction for Time-series Data (B-CIPIT), for learning from a sequence of time-series tasks. First, we develop a Bayesian LSTM based continual learning algorithm, which is capable of learning continually from a sequence of multi-variate time-series tasks, without catastrophically forgetting any representations. Second, we impute missing values in these time-series sequences, in a continual learning setting. We demonstrate and evaluate the robustness of the proposed algorithm on two real-world clinical time-series data sets, namely MIMIC-III [1] and PhysioNet Challenge 2012 [2]. Performance study results show the superiority of the proposed learning algorithm.
@inproceedings{guo_bayesian_2022,
	title = {Bayesian {Continual} {Imputation} and {Prediction} {For} {Irregularly} {Sampled} {Time} {Series} {Data}},
	issn = {2379-190X},
	doi = {10.1109/ICASSP43922.2022.9746342},
	abstract = {Learning from irregularly sampled, streaming, multi-variate time-series data with many missing values is a very challenging task. In this paper, we propose a Bayesian Continual Imputation and Prediction for Time-series Data (B-CIPIT), for learning from a sequence of time-series tasks. First, we develop a Bayesian LSTM based continual learning algorithm, which is capable of learning continually from a sequence of multi-variate time-series tasks, without catastrophically forgetting any representations. Second, we impute missing values in these time-series sequences, in a continual learning setting. We demonstrate and evaluate the robustness of the proposed algorithm on two real-world clinical time-series data sets, namely MIMIC-III [1] and PhysioNet Challenge 2012 [2]. Performance study results show the superiority of the proposed learning algorithm.},
	booktitle = {{ICASSP} 2022 - 2022 {IEEE} {International} {Conference} on {Acoustics}, {Speech} and {Signal} {Processing} ({ICASSP})},
	author = {Guo, Yang and Jun Poh, Jeanette Wen and Wong, Cheryl Sze Yin and Ramasamy, Savitha},
	month = may,
	year = {2022},
	keywords = {Bayesian Long Short-Term Memory, Continual Learning, Data models, MIMICs, Missing Data Imputation, Multi-variate Time Series, Prediction algorithms, Robustness, Signal processing, Signal processing algorithms, Time series analysis},
	pages = {4493--4497},
}

Downloads: 0