Recurrent Neural Networks (RNNs)
Resources
If you want to add more value to this section, we recommend checking out:
These resources will compliment this course, so we hope you enjoy!
Plan of Attack
- What we will learn in this section:
The Idea Behind Recurrent Neural Networks
| Supervised |
Artificial Neural Networks |
Used for Regression & Classification |
|
Convolutional Neural Networks |
Used for Computer Vision |
|
Recurrent Neural Networks |
Used for Time Series Analysis |
| Unsupervised |
Self-Organizing Maps |
Used for Feature Detection |
|
Deep Boltzmann Machines |
Used for Recommendation Systems |
|
AutoEncoders |
Used for Recommendation Systems |
Additional Reading
Sunspring (movie, 2016)
- Directed By Oscar Sharp
- Written By Benjamin
.png)
The Vanishing Gradient Problem

Wrec -> Recurrent Weight
Wrec ~ small -> Vanishing
Wrec ~ large -> Exploding
Solutions:
- Exploding Gradient
- Truncated Backpropagation
- Penalties
- Gradient Clipping
- Vanishing Gradient
- Weight Initialization
- Echo State Networks
- Long Short-Term Memory Networks (LSTMs)
Additional Reading
Untersuchungen zu dynamischen neuronalen Netzen
By Sepp (Josef) Hochreiter (1991)

Additional Reading
Learning Long-Term Dependencies with Gradient Descent is Difficult
By Yoshua Bengio et al. (1994)

Additional Reading
On The Difficulty of Training Recurrent Neural Networks
By Razvan Pascanu et al. (2013)

LSTMs
- Today:
- A bit of History
- LSTM Architecture
- Example Walkthrough
Reference:
Additional Reading
Long Short-Term Memory
By Sepp Hochreiter & Jurgen Schmidhuber (1997)

Additional Reading
Understanding LSTM Networks
By Christopher Olah (2015)

LSTM Practical Intuition
Reference:
Additional Reading
The Unreasonable Effectiveness of Recurrent Neural Networks
By Andrej Karpathy (2015)

Additional Reading
Visualizing and Understanding Recurrent Networks
By Andrej Karpathy et al. (2015)

LSTM Variations
Additional Reading
LSTM: A Search Space Odyssey
By Klaus Greff et al. (2015)
