An approach for output decoding of neural networks Strauß, Tobias 1984- VerfasserIn aut Labahn, Roger 1959- VerfasserIn aut Leifert, Gundram VerfasserIn aut Wustlich, Welf VerfasserIn aut text Text gw 2012 monographic XA-DE Rostock University of Rostock, Institute of Mathematics publisher 2012 Rostock Universitätsbibliothek distributor eng
electronic resource
remote
Computermedien
Online-Ressource
1 Online-Ressource (7 Seiten) We review two decoding methods for recurrent neural networks – based on the well known weighted Levenshtein distance and the connectionist temporal classification (CTC) – and we introduce a new method – the dynamic Levenstein distance (DynWL). We compare these three methods analytically in terms of time complexity and error performance. Although the approaches are different, there are deep connections between These ways of decoding. Finally, we test on the Arabic and French ICDAR data sets. Our experiments show that CTC yields the smallest error rates. Nevertheless, there are scenarios where DynWL is a good choice between performance and time complexity.<eng> Open Access Urheberrechtsschutz 1.0 Tobias Strauß, Roger Labahn, Gundram Leifert, Welf Wustlich 510 https://purl.uni-rostock.de/rosdok/id00005820 Preprints aus dem Institut für Mathematik 12/3 urn:nbn:de:gbv:28-rosdok_id00005820-5 10.18453/rosdok_id00005820 1621163060 rda DE-627 261010 1986180786 20261013T150735.0 Converted from MARCXML to MODS version 3.8 using MARC21slim2MODS3-8_XSLT1-0.xsl (Revision 1.174 20250328) ger