Inferential Machine Learning
Machine learning methods are increasingly being applied for automating prediction and classification tasks, for example customer recommender systems, spam detection or medical diagnosing. Recently there has also been an increased interest in methods that can explain why a machine learning system made a specific prediction or classification, typically referred to as Explainable Artificial Intelligence (XAI). For example that "patient X was diagnosed with disease D because of skin temperature B, blood level C etc". In this project we will attempt to take the field of XAI one step further by not making explanation for an individual prediction, but compile this to more overall explanations. For example that "smoking is associated with increased risk of disease D".
We will explore methods such as machine learning, XAI and potentially quantification of uncertainty. The project can be done using traditional machine learning methods or deep learning methods.
Learning outcome
Develop and evaluate techniques to do inferential machine learning.
Qualifications
Competence in machine learning.
Supervisors
References
Interpretable Machine Learning - A Guide for Making Black Box Models Explainable (Especially Chapter 8)
