Long Short Term Memory networks (LSTMs) are a type of recurrent neural network that can capture long term dependencies, which are frequently used for natural language modeling and speech recognition. This tutorial covers the conceptual basics of LSTMs and implements a basic LSTM in TensorFlow. The second part of the tutorial introduces the basics of TensorFlow, an Open Source software package used for implementing neural networks.

Taught by: Harini Suresh & Nick Locascio, MIT

Video:

• An introduction to LSTMs in TensorFlow (59:45)

Slides:

• An introduction to LSTMs in TensorFlow

Additional Resources:

• GitHub: iPython tutorials (code and data files, Jupyter Notebook)
• GitHub: iPython and TensorFlow setup