Emphasizing IML for explainability
IML is the paradigm of designing intelligent user interfaces to facilitate ML and AI algorithms with the help of user interactions. Using IML to steer the usage of ML systems to increase the trust of the end user has been an important research topic for the AI and HCI research community over the last few years. Many works of research literature recommend using IML to increase user engagement for AI systems. Recent Research Advances on Interactive Machine Learning by Jiang et al. (https://arxiv.org/abs/1811.04548) talks about some of the significant progress that has been made in the field of IML and how it is closely associated with the increasing trust and transparency of ML algorithms.
IML is another interesting approach that is used by the XAI community to explain ML models. Even in frameworks such as DALEX and Explainerdashboards, as covered in Chapter 9, Other Popular XAI Frameworks, providing interactive dashboards and web interfaces that...