(Not) Understanding Latin Poetic Style with Deep Learning

arXiv:2404.06150v1 Announce Type: new
Abstract: This article summarizes some mostly unsuccessful attempts to understand authorial style by examining the attention of various neural networks (LSTMs and CNNs) trained on a corpus of classical Latin verse that has been encoded to include sonic and metrical features. Carefully configured neural networks are shown to be extremely strong authorship classifiers, so it is hoped that they might therefore teach `traditional' readers something about how the authors differ in style. Sadly their reasoning is, …

abstract article arxiv attention cnns cs.cl deep learning features networks neural networks sonic style type understanding