2021 BEETL Competition: Advancing Transfer Learning for Subject Independence and Heterogenous EEG Data Sets

Xiaoxi Wei, A. Aldo Faisal, Moritz Grosse-Wentrup, Alexandre Gramfort, Sylvain Chevallier, Vinay Jayaram, Camille Jeunet, Stylianos Bakas, Siegfried Ludwig, Konstantinos Barmpas, Mehdi Bahri, Yannis Panagakis, Nikolaos Laskaris, Dimitrios A. Adamos, Stefanos Zafeiriou, William C. Duong, Stephen M. Gordon, Vernon J. Lawhern, Maciej Śliwowski, Vincent Rouanne, Piotr Tempczyk
Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track, PMLR 176:205-219, 2022.

Abstract

Transfer learning and meta-learning offer some of the most promising avenues to unlock the scalability of healthcare and consumer technologies driven by biosignal data. This is because regular machine learning methods cannot generalise well across human subjects and handle learning from different, heterogeneously collected data sets, thus limiting the scale of training data available. On the other hand, the many developments in transfer- and meta-learning fields would benefit significantly from a real-world benchmark with immediate practical application. Therefore, we pick electroencephalography (EEG) as an exemplar for all the things that make biosignal data analysis a hard problem. We design two transfer learning challenges around a. clinical diagnostics and b. neurotechnology. These two challenges are designed to probe algorithmic performance with all the challenges of biosignal data, such as low signal-to-noise ratios, major variability among subjects, differences in the data recording sessions and techniques, and even between the specific BCI tasks recorded in the dataset. Task 1 is centred on the field of medical diagnostics, addressing automatic sleep stage annotation across subjects. Task 2 is centred on Brain-Computer Interfacing (BCI), addressing motor imagery decoding across both subjects and data sets. The successful 2021 BEETL competition with its over 30 competing teams and its 3 winning entries brought attention to the potential of deep transfer learning and combinations of set theory and conventional machine learning techniques to overcome the challenges. The results set a new state-of-the-art for the real-world BEETL benchmarks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v176-wei22a, title = {2021 BEETL Competition: Advancing Transfer Learning for Subject Independence and Heterogenous EEG Data Sets}, author = {Wei, Xiaoxi and Faisal, A. Aldo and Grosse-Wentrup, Moritz and Gramfort, Alexandre and Chevallier, Sylvain and Jayaram, Vinay and Jeunet, Camille and Bakas, Stylianos and Ludwig, Siegfried and Barmpas, Konstantinos and Bahri, Mehdi and Panagakis, Yannis and Laskaris, Nikolaos and Adamos, Dimitrios A. and Zafeiriou, Stefanos and Duong, William C. and Gordon, Stephen M. and Lawhern, Vernon J. and {\'S}liwowski, Maciej and Rouanne, Vincent and Tempczyk, Piotr}, booktitle = {Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track}, pages = {205--219}, year = {2022}, editor = {Kiela, Douwe and Ciccone, Marco and Caputo, Barbara}, volume = {176}, series = {Proceedings of Machine Learning Research}, month = {06--14 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v176/wei22a/wei22a.pdf}, url = {https://proceedings.mlr.press/v176/wei22a.html}, abstract = {Transfer learning and meta-learning offer some of the most promising avenues to unlock the scalability of healthcare and consumer technologies driven by biosignal data. This is because regular machine learning methods cannot generalise well across human subjects and handle learning from different, heterogeneously collected data sets, thus limiting the scale of training data available. On the other hand, the many developments in transfer- and meta-learning fields would benefit significantly from a real-world benchmark with immediate practical application. Therefore, we pick electroencephalography (EEG) as an exemplar for all the things that make biosignal data analysis a hard problem. We design two transfer learning challenges around a. clinical diagnostics and b. neurotechnology. These two challenges are designed to probe algorithmic performance with all the challenges of biosignal data, such as low signal-to-noise ratios, major variability among subjects, differences in the data recording sessions and techniques, and even between the specific BCI tasks recorded in the dataset. Task 1 is centred on the field of medical diagnostics, addressing automatic sleep stage annotation across subjects. Task 2 is centred on Brain-Computer Interfacing (BCI), addressing motor imagery decoding across both subjects and data sets. The successful 2021 BEETL competition with its over 30 competing teams and its 3 winning entries brought attention to the potential of deep transfer learning and combinations of set theory and conventional machine learning techniques to overcome the challenges. The results set a new state-of-the-art for the real-world BEETL benchmarks.} }
Endnote
%0 Conference Paper %T 2021 BEETL Competition: Advancing Transfer Learning for Subject Independence and Heterogenous EEG Data Sets %A Xiaoxi Wei %A A. Aldo Faisal %A Moritz Grosse-Wentrup %A Alexandre Gramfort %A Sylvain Chevallier %A Vinay Jayaram %A Camille Jeunet %A Stylianos Bakas %A Siegfried Ludwig %A Konstantinos Barmpas %A Mehdi Bahri %A Yannis Panagakis %A Nikolaos Laskaris %A Dimitrios A. Adamos %A Stefanos Zafeiriou %A William C. Duong %A Stephen M. Gordon %A Vernon J. Lawhern %A Maciej Śliwowski %A Vincent Rouanne %A Piotr Tempczyk %B Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track %C Proceedings of Machine Learning Research %D 2022 %E Douwe Kiela %E Marco Ciccone %E Barbara Caputo %F pmlr-v176-wei22a %I PMLR %P 205--219 %U https://proceedings.mlr.press/v176/wei22a.html %V 176 %X Transfer learning and meta-learning offer some of the most promising avenues to unlock the scalability of healthcare and consumer technologies driven by biosignal data. This is because regular machine learning methods cannot generalise well across human subjects and handle learning from different, heterogeneously collected data sets, thus limiting the scale of training data available. On the other hand, the many developments in transfer- and meta-learning fields would benefit significantly from a real-world benchmark with immediate practical application. Therefore, we pick electroencephalography (EEG) as an exemplar for all the things that make biosignal data analysis a hard problem. We design two transfer learning challenges around a. clinical diagnostics and b. neurotechnology. These two challenges are designed to probe algorithmic performance with all the challenges of biosignal data, such as low signal-to-noise ratios, major variability among subjects, differences in the data recording sessions and techniques, and even between the specific BCI tasks recorded in the dataset. Task 1 is centred on the field of medical diagnostics, addressing automatic sleep stage annotation across subjects. Task 2 is centred on Brain-Computer Interfacing (BCI), addressing motor imagery decoding across both subjects and data sets. The successful 2021 BEETL competition with its over 30 competing teams and its 3 winning entries brought attention to the potential of deep transfer learning and combinations of set theory and conventional machine learning techniques to overcome the challenges. The results set a new state-of-the-art for the real-world BEETL benchmarks.
APA
Wei, X., Faisal, A.A., Grosse-Wentrup, M., Gramfort, A., Chevallier, S., Jayaram, V., Jeunet, C., Bakas, S., Ludwig, S., Barmpas, K., Bahri, M., Panagakis, Y., Laskaris, N., Adamos, D.A., Zafeiriou, S., Duong, W.C., Gordon, S.M., Lawhern, V.J., Śliwowski, M., Rouanne, V. & Tempczyk, P.. (2022). 2021 BEETL Competition: Advancing Transfer Learning for Subject Independence and Heterogenous EEG Data Sets. Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track, in Proceedings of Machine Learning Research 176:205-219 Available from https://proceedings.mlr.press/v176/wei22a.html.

Related Material