Algorithm-Level Deep Learning Techniques
The data-level deep learning techniques have problems very similar to classical ML techniques. Since deep learning algorithms are quite different from classical ML techniques, we’ll explore some algorithm-level techniques for addressing data imbalance in this chapter. These algorithm-level techniques won’t change the data but accommodate the model instead. This exploration might uncover new insights or methods to better handle imbalanced data.
This chapter will be on the same lines as Chapter 5, Cost-Sensitive Learning, extending the ideas to deep learning models. We will look at algorithm-level deep learning techniques to handle the imbalance in data. Generally, these techniques do not modify the training data and often require no pre-processing steps, offering the benefit of no increased training times or additional runtime hardware costs.
In this chapter, we will cover the following topics:
- Motivation for algorithm...
