[2024] Interactive Machine Learning: Visualization & Explainability
This course gives an introduction to the design of mixed-initiative systems.
Announcements
- The lecture will consist of a theoretical and a practical part.
- Students will work on a practical project throughout the semester, which will account for 50% of the final grade.
- The framework of the practical projects will be outlined during the first lecture (introduction).
- All materials and links will be shared on Moodel.
Announcement for interested domain scientists and collaborators: we are collecting proposals for project ideas through this form: external pagehttps://bit.ly/xaiml24-projects - feel free to submit project candidates.
Introduction
Interactive, mixed-initiative machine learning promises to combine the efficiency of automation with the effectiveness of humans for a collaborative decision-making and problem-solving process. This can be facilitated through co-adaptive visual interfaces.
The lecture consists of a theoretical and a practical part. The theoretical part will first introduce the foundations of information visualization design based on data charecteristics, e.g., high-dimensional, geo-spatial, relational, temporal, and textual data. In the second half of the theoretical lecture, we will discuss interaction techniques and explanation strategies to enable explainable machine learning with the tasks of understanding, diagnosing, and refining machine learning models.
The practical lecture part will cover the basics of back-end and front-end development and deployment, go deeper into D3.js and finally give more insight into explainable and interactive machine learning and data provence.
Learning Objectives
The goal of the course is to introduce techniques for interactive information visualization and to apply these on understanding, diagnosing, and refining machine learning models.
Schedule
The course is split into two parts, a theoretical and a practical part. Together these parts will cover the following topics:
- Foundations
- Back-end/front-end framework
- Visualization
- Interaction
- Explainability
- Outlook
Course Materials
Course materials will be made available on the Moodle platform.
Questions
To allow for an optimal flow of information, please ask your content-related questions on Moodle rather than via email. In this manner, your question and our answer are visible to everyone. Consequently, please read existing question-answer pairs before asking new questions.