In this chapter, we will discuss the main differences between machine learning and deep learning concepts.
Data Volume
Machine learning is suitable for different amounts of data, mainly for small-scale data. On the other hand, deep learning is more efficient when the amount of data increases rapidly. The following diagram illustrates how machine learning and deep learning work in terms of data volume.
Hardware Dependence
Contrary to traditional machine learning algorithms, deep learning algorithms are designed to rely heavily on high-end computers. Deep learning algorithms perform a large number of matrix multiplication operations, which require large-scale hardware support.
Feature Engineering
Feature engineering is the process of applying domain knowledge to specified features to reduce the complexity of the data and enable learning algorithms to recognize patterns.
For example, traditional machine learning models focus on pixels and other attributes needed in the feature engineering process. Deep learning algorithms focus on high-level features from data. This reduces the task of developing new feature extractors for each new problem.