SOM-CPC: Unsupervised Contrastive Learning with Self-Organizing Maps for Structured Representations of High-Rate Time Series

Iris A.M. Huijben, Arthur Andreas Nijdam, Sebastiaan Overeem, Merel M Van Gilst, Ruud Van Sloun
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:14132-14152, 2023.

Abstract

Continuous monitoring with an ever-increasing number of sensors has become ubiquitous across many application domains. However, acquired time series are typically high-dimensional and difficult to interpret. Expressive deep learning (DL) models have gained popularity for dimensionality reduction, but the resulting latent space often remains difficult to interpret. In this work we propose SOM-CPC, a model that visualizes data in an organized 2D manifold, while preserving higher-dimensional information. We address a largely unexplored and challenging set of scenarios comprising high-rate time series, and show on both synthetic and real-life data (physiological data and audio recordings) that SOM-CPC outperforms strong baselines like DL-based feature extraction, followed by conventional dimensionality reduction techniques, and models that jointly optimize a DL model and a Self-Organizing Map (SOM). SOM-CPC has great potential to acquire a better understanding of latent patterns in high-rate data streams.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-huijben23a, title = {{SOM}-{CPC}: Unsupervised Contrastive Learning with Self-Organizing Maps for Structured Representations of High-Rate Time Series}, author = {Huijben, Iris A.M. and Nijdam, Arthur Andreas and Overeem, Sebastiaan and Van Gilst, Merel M and Van Sloun, Ruud}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {14132--14152}, 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/huijben23a/huijben23a.pdf}, url = {https://proceedings.mlr.press/v202/huijben23a.html}, abstract = {Continuous monitoring with an ever-increasing number of sensors has become ubiquitous across many application domains. However, acquired time series are typically high-dimensional and difficult to interpret. Expressive deep learning (DL) models have gained popularity for dimensionality reduction, but the resulting latent space often remains difficult to interpret. In this work we propose SOM-CPC, a model that visualizes data in an organized 2D manifold, while preserving higher-dimensional information. We address a largely unexplored and challenging set of scenarios comprising high-rate time series, and show on both synthetic and real-life data (physiological data and audio recordings) that SOM-CPC outperforms strong baselines like DL-based feature extraction, followed by conventional dimensionality reduction techniques, and models that jointly optimize a DL model and a Self-Organizing Map (SOM). SOM-CPC has great potential to acquire a better understanding of latent patterns in high-rate data streams.} }
Endnote
%0 Conference Paper %T SOM-CPC: Unsupervised Contrastive Learning with Self-Organizing Maps for Structured Representations of High-Rate Time Series %A Iris A.M. Huijben %A Arthur Andreas Nijdam %A Sebastiaan Overeem %A Merel M Van Gilst %A Ruud Van Sloun %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-huijben23a %I PMLR %P 14132--14152 %U https://proceedings.mlr.press/v202/huijben23a.html %V 202 %X Continuous monitoring with an ever-increasing number of sensors has become ubiquitous across many application domains. However, acquired time series are typically high-dimensional and difficult to interpret. Expressive deep learning (DL) models have gained popularity for dimensionality reduction, but the resulting latent space often remains difficult to interpret. In this work we propose SOM-CPC, a model that visualizes data in an organized 2D manifold, while preserving higher-dimensional information. We address a largely unexplored and challenging set of scenarios comprising high-rate time series, and show on both synthetic and real-life data (physiological data and audio recordings) that SOM-CPC outperforms strong baselines like DL-based feature extraction, followed by conventional dimensionality reduction techniques, and models that jointly optimize a DL model and a Self-Organizing Map (SOM). SOM-CPC has great potential to acquire a better understanding of latent patterns in high-rate data streams.
APA
Huijben, I.A., Nijdam, A.A., Overeem, S., Van Gilst, M.M. & Van Sloun, R.. (2023). SOM-CPC: Unsupervised Contrastive Learning with Self-Organizing Maps for Structured Representations of High-Rate Time Series. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:14132-14152 Available from https://proceedings.mlr.press/v202/huijben23a.html.

Related Material