A new approach for cells in multidimensional recurrent neural networks Leifert, Gundram (VerfasserIn) Strauß, Tobias , 1984- (VerfasserIn) Labahn, Roger , 1959- (VerfasserIn) Wustlich, Welf (VerfasserIn) Text Text 2012 2012 eng electronic resourceremoteComputermedienOnline-Ressource1 Online-Ressource (22 Seiten) A recent approach for offline handwriting recognition is to use multidimensional recurrent neural networks (MDRNN) with connectionist temporal classification which has shown to yield very good results on several datasets. MDRNNs contain special units – multidimensional Long Short-Term Memory (MDLSTM) cells. These cells suffer from instability especially for higher dimensionality. We analyze the reasons for this effect and introduce several cells with better stability. We present a method to design stable multidimensional cells using the theory of linear shift invariant systems. The new cells are compared to MDLSTMs on the Arabic and French ICDAR datasets, where they yield better results.<eng> Open Access Urheberrechtsschutz 1.0 Gundram Leifert, Tobias Strauß, Roger Labahn, Welf Wustlich 510 https://purl.uni-rostock.de/rosdok/id00005819 urn:nbn:de:gbv:28-rosdok_id00005819-6 doi: 10.18453/rosdok_id00005819 oclc: 1621162786 ppn: (DE-627)1986180778