RNN with a recurrent output layer for learning of naturalness

Ján Dolinský, Hideyuki Takagi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The behavior of recurrent neural networks with a recurrent output layer (ROL) is described mathematically and it is shown that using ROL is not only advantageous, but is in fact crucial to obtaining satisfactory performance for the proposed naturalness learning. Conventional belief holds that employing ROL often substantially decreases the performance of a network or renders the network unstable, and ROL is consequently rarely used. The objective of this paper is to demonstrate that there are cases where it is necessary to use ROL. The concrete example shown models naturalness in handwritten letters.

Original languageEnglish
Title of host publicationNeural Information Processing - 14th International Conference, ICONIP 2007, Revised Selected Papers
Pages248-257
Number of pages10
EditionPART 1
DOIs
Publication statusPublished - 2008
Event14th International Conference on Neural Information Processing, ICONIP 2007 - Kitakyushu, Japan
Duration: Nov 13 2007Nov 16 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume4984 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other14th International Conference on Neural Information Processing, ICONIP 2007
CountryJapan
CityKitakyushu
Period11/13/0711/16/07

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'RNN with a recurrent output layer for learning of naturalness'. Together they form a unique fingerprint.

  • Cite this

    Dolinský, J., & Takagi, H. (2008). RNN with a recurrent output layer for learning of naturalness. In Neural Information Processing - 14th International Conference, ICONIP 2007, Revised Selected Papers (PART 1 ed., pp. 248-257). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4984 LNCS, No. PART 1). https://doi.org/10.1007/978-3-540-69158-7_27