Recurrent neural networks (RNNs) are capable of learning features and long
term dependencies from sequential and time-series data. The RNNs have a stack
of non-linear units where at least one connection between units forms a
directed cycle. A well-trained RNN can model any dynamical system; however,
training RNNs is mostly plagued by issues in learning long-term dependencies.
In this paper, we present a survey on RNNs and several new advances for
newcomers and professionals in the field. The fundamentals and recent advances
are explained and the research challenges are introduced.
%0 Generic
%1 salehinejad2017recent
%A Salehinejad, Hojjat
%A Sankar, Sharan
%A Barfett, Joseph
%A Colak, Errol
%A Valaee, Shahrokh
%D 2017
%K RNN seminar
%T Recent Advances in Recurrent Neural Networks
%U http://arxiv.org/abs/1801.01078
%X Recurrent neural networks (RNNs) are capable of learning features and long
term dependencies from sequential and time-series data. The RNNs have a stack
of non-linear units where at least one connection between units forms a
directed cycle. A well-trained RNN can model any dynamical system; however,
training RNNs is mostly plagued by issues in learning long-term dependencies.
In this paper, we present a survey on RNNs and several new advances for
newcomers and professionals in the field. The fundamentals and recent advances
are explained and the research challenges are introduced.
@misc{salehinejad2017recent,
abstract = {Recurrent neural networks (RNNs) are capable of learning features and long
term dependencies from sequential and time-series data. The RNNs have a stack
of non-linear units where at least one connection between units forms a
directed cycle. A well-trained RNN can model any dynamical system; however,
training RNNs is mostly plagued by issues in learning long-term dependencies.
In this paper, we present a survey on RNNs and several new advances for
newcomers and professionals in the field. The fundamentals and recent advances
are explained and the research challenges are introduced.},
added-at = {2018-02-26T11:14:33.000+0100},
author = {Salehinejad, Hojjat and Sankar, Sharan and Barfett, Joseph and Colak, Errol and Valaee, Shahrokh},
biburl = {https://www.bibsonomy.org/bibtex/2185eb3aee251610748fa588e3f068279/jk_itwm},
description = {Recent Advances in Recurrent Neural Networks},
interhash = {41ca950dd8759914915fee770a6484d8},
intrahash = {185eb3aee251610748fa588e3f068279},
keywords = {RNN seminar},
note = {cite arxiv:1801.01078Comment: arXiv admin note: text overlap with arXiv:1602.04335},
timestamp = {2018-02-26T11:14:33.000+0100},
title = {Recent Advances in Recurrent Neural Networks},
url = {http://arxiv.org/abs/1801.01078},
year = 2017
}