Long short-term memory

Long short-term memory (LSTM) is a recurrent neural network (RNN) architecture (an artificial neural network) proposed in 1997 by Sepp Hochreiter and Jürgen Schmidhuber.

Documentation

Show and Tell - A Neural Image Caption Generator

https://arxiv.org/pdf/1411.4555v2.pdf

Long-term Recurrent Convolutional Networks for Visual Recognition and Description

Easy way to combine CNN + LSTM? (e.g. LRCN network)

https://arxiv.org/pdf/1411.4389v3.pdf

See also

• RNN
• BPTT
• Vanishing Gradients
• Autoregressive integrated moving average (ARIMA)

Favorite site

• Wikipedia (en) Long short-term memory
• [추천] Understanding LSTM Networks
  • Kor ver: Recurrent Neural Network (RNN) 과 LSTM ¹
• Team AI Korea - Recurrent Neural Network (RNN) Tutorial - Part 1
• 초보자를 위한 RNNs과 LSTM 가이드
• 머신러닝레볼루션: RNN과 LSTM - 쫄지말자 딥러닝
• ARIMA vs LSTM - Forecasting Weekly Hotel Cancellations

References