Elevator Pitch
Long Short-Term Memory (LSTM) is a Recurrent Neural Network (RNN) architecture that remembers values over arbitrary intervals. LSTMs have been known to have achieved state of the art in many sequence classification problems. In this talk, I’ll show how to write an LSTM using TensorFlow’s Python API.
Description
Recurrent Neural Networks (RNNs) are one of the most important areas in Deep Learning. RNNs enable numerous exciting applications in speech recognition, audio synthesis, machine translation, natural language understanding, and many other areas.
Long Short-Term Memory (LSTM) is a Recurrent Neural Network (RNN) architecture that remembers values over long intervals. LSTMs have been known to have achieved state of the art performance in many sequence classification problems. In this talk, I’ll cover how to write an LSTM using TensorFlow’s Python API for natural language understanding.
This is going to be a code-heavy talk where I will implement the LSTM model and explain the math behind it step-by-step. In short, it will cover -
- Understanding how the math behind LSTM architecture works in case of sequence classification
- Writing an LSTM model using TensorFlow for sentiment classification of variable length English language sentences
- Explaining how dropout works in an LSTM architecture
Notes
I’m currently working on one of the projects based on Long Short-Term Memory (LSTM) at Booking.com. Here, I aim to share my knowledge and learning from this experience while working through a basic yet powerful application of LSTM in sequence modeling.