EPSTE decomposes MEG time series into geometric symbolic tokens and uses an attention RNN to predict surrogate-validated transfer entropy, recovering directed structure more accurately than a standard symbolic baseline on AAL90-parcellated data.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Embedded Polygon Symbolic Transfer Entropy (EPSTE): A Geometric Token and Deep Learning Approach to Estimating Transfer Entropy in Neuroimaging Time Series
EPSTE decomposes MEG time series into geometric symbolic tokens and uses an attention RNN to predict surrogate-validated transfer entropy, recovering directed structure more accurately than a standard symbolic baseline on AAL90-parcellated data.