Feature Programming for Multivariate Time Series Prediction

Alex Daniel Reneau, Jerry Yao-Chieh Hu, Ammar Gilani, Han Liu
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:29009-29029, 2023.

Abstract

We introduce the concept of programmable feature engineering for time series modeling and propose a feature programming framework. This framework generates large amounts of predictive features for noisy multivariate time series while allowing users to incorporate their inductive bias with minimal effort. The key motivation of our framework is to view any multivariate time series as a cumulative sum of fine-grained trajectory increments, with each increment governed by a novel spin-gas dynamical Ising model. This fine-grained perspective motivates the development of a parsimonious set of operators that summarize multivariate time series in an abstract fashion, serving as the foundation for large-scale automated feature engineering. Numerically, we validate the efficacy of our method on several synthetic and real-world noisy time series datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-reneau23a, title = {Feature Programming for Multivariate Time Series Prediction}, author = {Reneau, Alex Daniel and Hu, Jerry Yao-Chieh and Gilani, Ammar and Liu, Han}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {29009--29029}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/reneau23a/reneau23a.pdf}, url = {https://proceedings.mlr.press/v202/reneau23a.html}, abstract = {We introduce the concept of programmable feature engineering for time series modeling and propose a feature programming framework. This framework generates large amounts of predictive features for noisy multivariate time series while allowing users to incorporate their inductive bias with minimal effort. The key motivation of our framework is to view any multivariate time series as a cumulative sum of fine-grained trajectory increments, with each increment governed by a novel spin-gas dynamical Ising model. This fine-grained perspective motivates the development of a parsimonious set of operators that summarize multivariate time series in an abstract fashion, serving as the foundation for large-scale automated feature engineering. Numerically, we validate the efficacy of our method on several synthetic and real-world noisy time series datasets.} }
Endnote
%0 Conference Paper %T Feature Programming for Multivariate Time Series Prediction %A Alex Daniel Reneau %A Jerry Yao-Chieh Hu %A Ammar Gilani %A Han Liu %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-reneau23a %I PMLR %P 29009--29029 %U https://proceedings.mlr.press/v202/reneau23a.html %V 202 %X We introduce the concept of programmable feature engineering for time series modeling and propose a feature programming framework. This framework generates large amounts of predictive features for noisy multivariate time series while allowing users to incorporate their inductive bias with minimal effort. The key motivation of our framework is to view any multivariate time series as a cumulative sum of fine-grained trajectory increments, with each increment governed by a novel spin-gas dynamical Ising model. This fine-grained perspective motivates the development of a parsimonious set of operators that summarize multivariate time series in an abstract fashion, serving as the foundation for large-scale automated feature engineering. Numerically, we validate the efficacy of our method on several synthetic and real-world noisy time series datasets.
APA
Reneau, A.D., Hu, J.Y., Gilani, A. & Liu, H.. (2023). Feature Programming for Multivariate Time Series Prediction. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:29009-29029 Available from https://proceedings.mlr.press/v202/reneau23a.html.

Related Material