In order to improve the performance of lstm model in deep learning, we can increase the depth of lstm networks. There are two types of models to increase the depth of lsm networks. They are:
- Nested lstm
To learn nested lstm, you can read this tutorial:
Understand Nested LSTM Network: A Beginner Guide – LSTM Network Tutorial
- Stacked lstm
In this tutorial, we will introduce the stacked lstm for deep learning beginners.
What is stacked lstm?
The structure of stacked lstm looks like:
We can stack LSTM layer one by one to increase the depth of LSTM netwoks.
We should notice the Dense layer, which can use a non-linear transformation to convert the lstm output to final output.
To understand Dense layer, you can read this tutorial:
Understand Dense Layer (Fully Connected Layer) in Neural Networks