RECORD
Self-supervised representation learning from electroencephalography signals
Machine Learning · Submitted 13 Nov 2019
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim As demonstrated on a clinically relevant task (sleep scoring) and with two electroencephalography datasets, our approach outperforms a purely supervised approach in low data regimes, while capturing important physiological information without any access to labels.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
The supervised learning paradigm is limited by the cost - and sometimes the impracticality - of data collection and labeling in multiple domains. Self-supervised learning, a paradigm which exploits the structure of unlabeled data to create learning problems that can be solved with standard supervised approaches, has shown great promise as a pretraining or feature learning approach in fields like computer vision and time series processing. In this work, we present self-supervision strategies that can be used to learn informative representations from multivariate time series. One successful approach relies on predicting whether time windows are sampled from the same temporal context or not. As demonstrated on a clinically relevant task (sleep scoring) and with two electroencephalography datasets, our approach outperforms a purely supervised approach in low data regimes, while capturing important physiological information without any access to labels.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | The supervised learning paradigm is limited by the cost - and sometimes the impracticality - of data collection and labeling in multiple domains. | background 0.97 | 0.05 | 0.02 | asserted 1.00 |
| s1 | Self-supervised learning, a paradigm which exploits the structure of unlabeled data to create learning problems that can be solved with standard supervised approaches, has shown great promise as a pretraining or feature learning approach in fields like computer vision and time series processing. | background 1.00 | 0.05 | 0.04 | asserted 0.98 |
| s2 | In this work, we present self-supervision strategies that can be used to learn informative representations from multivariate time series. | contribution 1.00 | 0.86 | 0.07 | asserted 0.77 |
| s3 | One successful approach relies on predicting whether time windows are sampled from the same temporal context or not. | method 0.64 | 0.12 | 0.04 | asserted 0.98 |
| s4 | As demonstrated on a clinically relevant task (sleep scoring) and with two electroencephalography datasets, our approach outperforms a purely supervised approach in low data regimes, while capturing important physiological information without any access to labels. | result 1.00 | 0.96 | 0.91 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
The supervised learning paradigm is limited by the cost - and sometimes the impracticality - of data collection and labeling in multiple domains. Self-supervised learning, a paradigm which exploits the structure of unlabeled data to create learning problems that can be solved with standard supervised approaches, has shown great promise as a pretraining or feature learning approach in fields like computer vision and time series processing. In this work, we present self-supervision strategies that can be used to learn informative representations from multivariate time series. One successful approach relies on predicting whether time windows are sampled from the same temporal context or not. As demonstrated on a clinically relevant task (sleep scoring) and with two electroencephalography datasets, our approach outperforms a purely supervised approach in low data regimes, while capturing important physiological information without any access to labels.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
Papers whose main result relates to this one's
States the same result 1 paper
-
We demonstrate the benefit of self-supervised learning approaches on EEG data.
Uncovering the structure of clinical EEG signals with self-supervised learning · 0.98
What it cites · 1 sentence without a confident label
Uses the method or tool of 2 papers
-
To further explore the features learned with SSL, we project the 100-dimension embeddings obtained on the labeled Sleep EDF dataset to two dimensions using UMAP [ref] .
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction · Experiment 3: SSL models learn physiologically meaningful features · 1.00
-
The Adam optimizer [ref] with $β_1$=0.9, $β_2$=0.999 and learning rate 0.001 is used, while the batch size is 256.
Adam: A Method for Stochastic Optimization · Model architecture · 1.00
Cites as background 1 paper
-
With SSL, the structure of the data is used to turn an unsupervised learning problem into a supervised learning problem, called a “pretext task” [ref] .
Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey · Introduction · 0.98
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.