Deep learning is a subtopic within machine learning that seeks to automate the feature selection process in addition to the learning, thus attempting to eliminate human intervention at an earlier stage of the process than “non-deep” machine learning. As classical machine learning relies on humans to provide the features used to train a model, it is usually more suited to structured problems where feature engineering is domain specific and benefits from subject matter knowledge of the data. On the other hand, deep learning is better suited to unstructured data that would be difficult for humans to use their knowledge to extract features, such as text or image data.
Deep Learning often refers to algorithms called Neural Networks. These attempt to mimic organic systems, i.e. the brain, and the term ‘deep’ refers to the Neural Network having a ‘stack’, or multiple layers. As these algorithms often require terabytes of data in order to learn from the training data, Deep Learning is considered a more modern technology compared to many others within Machine Learning. The number of multi-dimensional calculations required for training Deep Learning models mean that only top specification computers can be relied on, and until the past few decades, computational power was a major constraint in being able to apply these algorithms in many business settings.