This book offers a complete framework for classifying and transcribing sequential data with recurrent neural networks. It uses state-of-the-art results in speech and handwriting recognition to show the framework in action.For example, sequences of handwritten digits must be presegmented into individual characters before they can be recognised by convolutional networks ( LeCun et al., 1998a). ... This provides an intuitive solution to the problem of simultaneously assimilating context from all directions. ... In two dimensions, this structure has been successfully used to evaluate positions in the board game a#39;Goa#39; (Wu andanbsp;...
|Title||:||Supervised Sequence Labelling with Recurrent Neural Networks|
|Publisher||:||Springer Science & Business Media - 2012-02-06|