Explainability in machine learning

Using state-of-the-art deep learning techniques, Dr Sen Wang’s research focuses on developing interpretable deep models in many data-rich fields to provide insights into the data, variables and decision points used to make a recommendation.

In health-related projects, Dr Wang strives to improve the performance of medical prediction, in areas such as illness severity prediction in Intensive Care Units, and addressing explainability issues raised by the ‘black-box’ models.

Another ML application is in the material sciences, where the team is seeking to discern and discover new compositions and properties of existing alloys by exploiting explainability using novel deep-learning approaches.