Understanding LSTM -- a tutorial into Long Short-Term Memory Recurrent\n Neural Networks

Long Short-Term Memory Recurrent Neural Networks (LSTM-RNN) are one of the\nmost powerful dynamic classifiers publicly known. The network itself and the\nrelated learning algorithms are reasonably well documented to get an idea how\nit works. This paper will shed more light into understanding how LSTM-RNNs\nevolved and why they work impressively well, focusing on the early,\nground-breaking publications. We significantly improved documentation and fixed\na number of errors and inconsistencies that accumulated in previous\npublications. To support understanding we as well revised and unified the\nnotation used.\n