LSTM implementation with theano (by Christian Herta)

LSTM for Deep Learning using theano with python code (by Pierre Luc Carrier and Kyunghyun Cho)

RNN/LSTM lib (by Alex Graves)

lstm-g

lstm-g-hardcoder

architecture-free neural network library for node.js

  • A. Graves, N. Jaitly, A. Mohamed. Hybrid Speech Recognition with Deep Bidirectional LSTM. ASRU 2013, Olomouc, Czech Republic.
  • Graves, Alex. Supervised sequence labelling with recurrent neural networks. Vol. 385. Springer, 2012.
  • Derek D. Monner , James A. Reggia. A generalized LSTM-like training algorithm for second-order recurrent neural networks. Neural Networks,  Jan; 25(1): 70–83, 2012. (here or at NN). 
  • J. A. Pérez-Ortiz, F. A. Gers, D. Eck, and J. Schmidhuber. Kalman filters improve LSTM network performance in problems unsolvable by traditional recurrent nets. Neural Networks, 16(2):241-250, 2002. (Gzipped PostScript, 23 pages, 122547 bytes) (PDF, 309253 bytes)
  • J. Schmidhuber, F. A. Gers, and D. Eck. Learning nonregular languages: A comparison of simple recurrent networks and LSTM. Neural Computation, 14(9):2039-2041, 2002. (Gzipped PostScript, 5 pages, 62797 bytes) (PDF, 71524 bytes)
  • F. A. Gers and J. Schmidhuber. LSTM recurrent networks learn simple context free and context sensitive languages. IEEE Transactions on Neural Networks, 12(6):1333-1340, 2001. (Gzipped PostScript, 14 pages, 130308 bytes) (PDF, 330683 bytes)
  • F. A. Gers. Long Short-Term Memory in Recurrent Neural Networks. PhD thesis, Department of Computer Science, Swiss Federal Institute of Technology, Lausanne, EPFL, Switzerland, 2001. (Gzipped PostScript, 102 pages, 485269 bytes) (PDF, 1300159 bytes)
  • F. A. Gers, J. Schmidhuber, and F. Cummins. Learning to forget: Continual prediction with LSTM. Neural Computation, 12(10):2451-2471, 2000. (Gzipped PostScript, 20 pages, 144674 bytes) (PDF, 358394 bytes)
  • F. A. Gers, J. Schmidhuber, and N. Schraudolph. Learning precise timing with LSTM recurrent networks. Journal of Machine Learning Research (JMLR), 3:115-143, 2002. (Gzipped PostScript, 29 pages, 172717 bytes) (PDF, 474948 bytes)
  • F. A. Gers, J. A. Pérez-Ortizand, D. Eck, and J. Schmidhuber. Learning context sensitive languages with LSTM trained with kalman filters. In J. Dorronsoro, editor, Proc. ICANN 2002, Int. Conf. on Artificial Neural Networks, pages 655-660, Berlin, 2002. Springer. (Gzipped PostScript, 6 pages, 46179 bytes) (PDF, 132140 bytes)
  • J. A. Pérez-Ortizand, J. Schmidhuber, F. A. Gers, and D. Eck. Improving long-term online prediction with decoupled extended kalman filters. In J. Dorronsoro, editor, Proc. ICANN 2002, Int. Conf. on Artificial Neural Networks, pages 1055-1060, Berlin, 2002. Springer. (Gzipped PostScript, 6 pages, 48964 bytes) (PDF, 151782 bytes)
  • F. A. Gers, D. Eck, and J. Schmidhuber. Applying LSTM to time series predictable through time-window approaches. In Proc. ICANN 2001, Int. Conf. on Artificial Neural Networks, Vienna, Austria, 2001. IEE, London. (Gzipped PostScript, 9 pages, 71227 bytes) (PDF, 181609 bytes)
  • F. A. Gers and J. Schmidhuber. Long Short-Term Memory learns context free and context sensitive languages. In Kurkova et./ al./, editor, Proceedings of the ICANNGA 2001 Conference, volume 1, pages 134-137, Wien,NY, 2001. Springer. (Gzipped PostScript, 4 pages, 54412 bytes) (PDF, 155749 bytes)
  • F. A. Gers and J. Schmidhuber. Neural processing of complex continual input streams. In Proc. IJCNN'2000, Int. Joint Conf. on Neural Networks, Como, Italy, 2000. (Gzipped PostScript, 6 pages, 68548 bytes) (PDF, 194541 bytes)
  • F. A. Gers and J. Schmidhuber. Recurrent nets that time and count. In Proc. IJCNN'2000, Int. Joint Conf. on Neural Networks, Como, Italy, 2000. (Gzipped PostScript, 6 pages, 73488 bytes) (PDF, 206047 bytes)
  • Fred Cummins, Felix Gers, and Jürgen Schmidhuber. Language identification from prosody without explicit features. In Proceedings of EUROSPEECH'99, volume 1, pages 371-374, 1999. (Gzipped PostScript, 6 pages, 50243 bytes) (PDF, 121554 bytes)
  • F. A. Gers, J. Schmidhuber, and F. Cummins. Learning to forget: Continual prediction with LSTM. In Proc. ICANN'99, Int. Conf. on Artificial Neural Networks, volume 2, pages 850-855, Edinburgh, Scotland, 1999. IEE, London. (Gzipped PostScript, 7 pages, 102808 bytes) (PDF, 283530 bytes)