A Technical Note on the Architectural Effects on Maximum Dependency Lengths of Recurrent Neural Networks

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kent, Jonathan S., Murray, Michael M.
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910573289013248
author Kent, Jonathan S.
Murray, Michael M.
author_facet Kent, Jonathan S.
Murray, Michael M.
contents This work proposes a methodology for determining the maximum dependency length of a recurrent neural network (RNN), and then studies the effects of architectural changes, including the number and neuron count of layers, on the maximum dependency lengths of traditional RNN, gated recurrent unit (GRU), and long-short term memory (LSTM) models.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11946
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Technical Note on the Architectural Effects on Maximum Dependency Lengths of Recurrent Neural Networks
Kent, Jonathan S.
Murray, Michael M.
Neural and Evolutionary Computing
I.2.6
This work proposes a methodology for determining the maximum dependency length of a recurrent neural network (RNN), and then studies the effects of architectural changes, including the number and neuron count of layers, on the maximum dependency lengths of traditional RNN, gated recurrent unit (GRU), and long-short term memory (LSTM) models.
title A Technical Note on the Architectural Effects on Maximum Dependency Lengths of Recurrent Neural Networks
topic Neural and Evolutionary Computing
I.2.6
url https://arxiv.org/abs/2408.11946