ABSTRACT
This chapter provides a brief introduction to machine learning and sketch the development from knowledge-based systems to deep representations. It discusses machine learning approaches based on random forests and deep convolutional neural networks as well as their application to the analysis of whole-slide histopathology sections as a basis to tissue phenomics. Machine learning is a branch of computer science attempting to create systems for inference and prediction that can learn from data without requiring the user to define explicit rules. An interesting idea to compensate for the weakness of machine learning approaches with respect to a lack of training data is to team knowledge-driven and data-driven representations. Independently of the chosen machine learning approach, a common good practice is also to normalize the input data that, in the context of the analysis of histopathology images, typically translates into stain normalization.
