An approach for output decoding of neural networks

10.18453/rosdok_id00005820 Strauß, Tobias Tobias Strauß http://d-nb.info/gnd/1136294554 https://orcid.org/0000-0002-5344-4188 Labahn, Roger Roger Labahn http://d-nb.info/gnd/170999645 https://orcid.org/0000-0003-1901-9644 Leifert, Gundram Gundram Leifert Wustlich, Welf Welf Wustlich An approach for output decoding of neural networks University of Rostock, Institute of Mathematics 2012 510 Mathematics 2012 en https://purl.uni-rostock.de/rosdok/id00005820 urn:nbn:de:gbv:28-rosdok_id00005820-5 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.