‘Data is useless without data science’, Pries writes in his dissertation. ‘This rapidly developing field aims to extract knowledge and understanding from any kind of data. It gives meaning to the bits and bytes in this world. In an age where we have an abundance of data, techniques from data science have had many success stories. Understanding what the data tells us is extremely useful for gaining insights and making predictions.’
The focus of many data scientists is on predictive methods for practical applications. Classification and regression techniques are able to automatically learn from data in order to make predictions about unseen data. In his dissertation, Pries questioned commonly used techniques and developed better alternatives. His research ‘gives practitioners the means to gain accurate insights and draw meaningful conclusions’.
Face generators
Pries’ dissertation consists of topics in the fields of artificial intelligence, machine learning, statistics, and data analysis. The first part is about face generators and active learning. An example of a face generator is the website thispersondoesnotexist.com, were a new human face is generated every time the site is refreshed. This face does not come from someone’s personal photo album. Instead, a model is trained to create new faces by learning from real images. Pries evaluated a face generator with a humanlike approach and performed a pioneering study to improve labeling of pairwise distance datasets that can be used to advance face recognition and likeness methods.
Analysis and statistics
The other part of the dissertation is about benchmarking binary classification methods, where Pries introduces a new baseline approach. Also he looked at two important subjects in data analysis and statistics: accurately quantifying how dependent one variable is on another variable, and determining how important a feature is for predicting a target variable.
Header photo: Shutterstock
