MAchine learning for environmental TIme Series – MATS

Task 1 deals with time series classification through the integration of localization information (task 1.1) and the creation of links between shapelet models and convolutional neural networks.
Task 2 focuses on time series modelling.
Task 3 tackles weakly supervised learning through semi-supervised approaches (task 3.1), metric learning (task 3.2) and domain adaptation (task 3.3).
Task 4 is an application task that aims at showcasing methodological developments from other tasks in real-worls settings.

[T1] The integration of temporal localization information has been tackled in [1] for Random Shapelet models. On the links between Shapelet modelss and CNNs, we have proposed (i) an adversarial regularization framework to help learn interpretable featuresin Learned-Shapelet-like models [2] and we have studied (ii) the use of hybrid approaches (based on both extraction learning of shapelets) in order to accelerate the learning process [3].
[T2] In the context of task 2, we have developped a clustering model for time series that operates in continuous time.
[T3] Our work on metricshas focused on the joint alignment of temporal axes and feature spaces. Moreover, recent works by members of the project that focus on optimal transport [6,7,8] will be used as a basis for the development of domain adaptation methods.
[T4] One can highlight the work by Emilien Alvarez, on questions related to habitat mapping from UAV data [9], as well as the collaboration with Marc Russwurm, TU Munich PhD student, on questions related to the early classificationof time series for agriculture monitoring [10].

Not relevant yet.

[0] Romain Tavenard et al.. Tslearn, A Machine Learning Toolkit for Time Series Data. In Journal of Machine Learning Research, vol. 21, pp. 1 - 6, 2020.
[1] Maël Guillemé, Simon Malinowski, Romain Tavenard, Xavier Renard. Localized Random Shapelets. In Proceedings of the International Workshop on Advanced Analysis and Learning on Temporal Data, Wurzburg, Germany, 2019.
[2] Yichang Wang, Rémi Emonet, Elisa Fromont, Simon Malinowski, Romain Tavenard. Adversarial Regularization for Explainable-by-Design Time Series Classification. In Proceedings of ICTAI 2020, Greece, 2020.
[3] David Guijo-Rubio, Pedro Gutiérrez, Romain Tavenard, Anthony Bagnall. A Hybrid Approach to Time Series Classification with Shapelets. In Proceedings of the Intelligent Data Engineering and Automated Learning -- IDEAL, Manchester, United Kingdom, 2019.
[6] Titouan Vayer, Laetitia Chapel, Rémi Flamary, Romain Tavenard, Nicolas Courty. Optimal Transport for structured data with application on graphs. In Proceedings of the ICML 2019 - 36th International Conference on Machine Learning, Long Beach, United States, 2019.
[7] Titouan Vayer, Rémi Flamary, Romain Tavenard, Laetitia Chapel, Nicolas Courty. Sliced Gromov-Wasserstein. In Proceedings of the NeurIPS 2019 - Thirty-third Conference on Neural Information Processing Systems, Vancouver, Canada, 2019.
[8] Titouan Vayer, Laetitia Chapel, Rémi Flamary, Romain Tavenard, Nicolas Courty. Fused Gromov-Wasserstein Distance for Structured Objects. In Algorithms, vol. 13, no 9, p. 212, 2020.
[9] Emilien Alvarez-Vanhard, Thomas Houet, Cendrine Mony, Lucie Lecoq, Thomas Corpetti. Can UAVs fill the gap between in situ surveys and satellites for habitat mapping? Remote Sensing of Environment, Elsevier, 2020, 243.
[10] Marc Rußwurm, Romain Tavenard, Sébastien Lefèvre, Marco Körner. Early Classification for Agricultural Monitoring from Satellite Time Series. In Proceedings of the AI for Social Good Workshop at ICML, Long Beach, United States, 2019.

A huge trend in recent earth observation missions is to target high temporal and spatial resolutions (\emph{e.g.} SENTINEL-2 mission by ESA).
Data resulting from these missions can then be used for fine-grained studies in many applications.
In this project we will focus on three key environmental issues: agricultural practices and their impact, forest preservation and air quality monitoring.

Based on identified key requirements for these application settings, MATS project will feature a complete rethinking of the literature in machine learning for time series, with a focus on large-scale methods that could operate even when little supervised information is available.
In more details, MATS will introduce new paradigms in large-scale time series classification, spatio-temporal modelling and weakly supervised approaches for time series.
Proposed methods will cover a wide range of machine learning problems including domain adaptation, clustering, metric learning and (semi-)supervised classification, for which dedicated methodology is lacking when time series data is at stake.
Methods developed in the project will be made available to the scientific community as well as to practitioners through an open-source toolbox in order to help dissemination to a wide range of application areas.
Moreover, the application settings considered in the project will be used to showcase benefits offered by methodologies developed in MATS in terms of time series analysis.