TY  - BOOK
AU  - Leifert, Gundram
AU  - Strauß, Tobias
AU  - Labahn, Roger
AU  - Wustlich, Welf
PY  - 2012
DA  - 2012//
TI  - A  new approach for cells in multidimensional recurrent neural networks
T3  - Preprints aus dem Institut für Mathematik
CY  - Germany; Rostock
AB  - 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>
UR  - https://purl.uni-rostock.de/rosdok/id00005819
UR  - https://doi.org/10.18453/rosdok_id00005819
DO  - 10.18453/rosdok_id00005819
LA  - English
N1  - Gundram Leifert, Tobias Strauß, Roger Labahn, Welf Wustlich
ID  - 1986180778
ER  -