Reproduce “Tokenizing Single-Channel EEG with Time-Frequency Motif Learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jathurshan Pradeepkumar, Xihao Piao, Zheng Chen, Jimeng Sun
TFM-Tokenizer turns EEG channels into discrete motif tokens by combining raw-signal and spectrogram features, quantizing segments with a learned codebook, and feeding tokens to downstream models. On TUEV six-class event classification after pretraining on four EEG datasets, it reports Cohen’s kappa of 0.6189 ± 0.0302, versus 0.5588 ± 0.0273 for CBraMod. Device transfer was tested only after fine-tuning on one ear-EEG dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.