Summary of Are Eeg Sequences Time Series? Eeg Classification with Time Series Models and Joint Subject Training, by Johannes Burchert et al.
Are EEG Sequences Time Series? EEG Classification with Time Series Models and Joint Subject Training
by Johannes Burchert, Thorben Werner, Vijaya Krishna Yalavarthi, Diego Coello de Portugal, Maximilian Stubbemann, Lars Schmidt-Thieme
First submitted to arxiv on: 10 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Signal Processing (eess.SP)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper systematically studies the differences between EEG classification models and generic time series classification models. It presents three different model setups to deal with EEG data from different subjects: subject-specific models (most common in EEG literature), subject-agnostic models, and subject-conditional models. The results show that off-the-shelf time series classification models trained per subject perform close to EEG classification models, but don’t quite reach the performance of domain-specific modeling. Additionally, the paper combines time-series models with subject embeddings to train one joint subject-conditional classifier on all subjects, which outperforms dedicated EEG methods in 2 out of 3 datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary EEG data analysis relies heavily on preprocessing, and machine learners often treat it like any other time series data. However, many models developed for EEG classification have unique layer types and architectures. This paper explores the differences between EEG classification models and generic time series classification models. It presents three different approaches: learning separate models per subject, one model for all subjects, or a combination of both. The results show that off-the-shelf time series models can perform well on EEG data, but don’t quite match the performance of specialized EEG models. |
Keywords
* Artificial intelligence * Classification * Time series
