Machine learning has great potential for improving products, processes and research. However, the width of the applications of machine learning techniques is narrowed by the lack of explanations of the methods used. The book by Molnar introduces the reader to interpretability of machine learning.
Within this seminar, you will learn about simple, interpretable models such as decision trees, decision rules and linear regression; moreover, you have the change to get to learn about state-of-the-art literature and recent preogresses. The focus of the book is on model-agnostic methods for interpreting black box models such as feature importance and accumulated local effects, and explaining individual predictions. In addition, the book presents methods specific to deep neural networks and evaluation of the interpretability of the methods.
All interpretation methods are explained in-depth and discussed critically. How do they work under the hood? What are their strengths and weaknesses? How can their outputs be interpreted? This book will enable you to select and correctly apply the interpretation method that is most suitable for your machine learning project. Reading the book is recommended for machine learning practitioners, data scientists, statisticians, and anyone else interested in making machine learning models interpretable.
</p>
<p>
<b> Literature</b>
</p>
<p>
This seminar relies on the book <b>Interpretable Machine Learning - A Guide for Making Black Box Models Explainable</b> by Christoph Molnar. This book is available for free here <ahref="https://christophm.github.io/interpretable-ml-book/">https://christophm.github.io/interpretable-ml-book/</a>. Please note that a single person wrote this book and thus probably contains some errors. So finding alternative sources is even more significant in this seminar.
Students enrolled in this Pro-Seminar will send their favorite topics (possibly with priorities) to <ahref="mailto:simon.kluettermann@cs.tu-dortmund.de?cc=chiara.balestra@cs.tu-dortmund.de">Simon Klüttermann</a> until <b>07.10.2022</b>. We will assign topics based on your choices in the following week. We will organize a short meeting to answer your questions and present before the deadline.
You will also be assigned a supervisor who will help with the questions you might have on the specific topic. For general questions, you can also always write <ahref="mailto:simon.kluettermann@cs.tu-dortmund.de?cc=chiara.balestra@cs.tu-dortmund.de">Simon Klüttermann</a>.
<b>We will provide an internal presentation course; hence you dont need to take the one offered by the faculty</b>. Also, we will hold the course in <b>English</b>.
The final talks will be scheduled over 1-3 days in the <b>second half of January</b> (each talk: 25/ 30 minutes). A written report about your topic before the deadline.
<h3>Goals and Criteria for a successful seminar</h3>
<p>
In the Pro-Seminar, you will learn how to work yourself into a topic, research related literature, and answer questions to this topic. To this end, you need to read the subchapters assigned to you and use different sources to verify and extend the statements made. Also, by listening and engaging with the other presentations, you will get a comprehensive understanding of interpretable machine learning methods.
You will learn to address constructive criticism on research topics. To train this, we will assign to you two students' reports to critically comment until the end of the semester. You need to participate in every part of this seminar to pass it.
