Introduction to Machine Learning For Engineering Research (ML4ER)

“Introduction to Machine Learning For Engineering Research” (oe, ML4ER) is a semester long course taught in the Spring and Fall semesters to serve as an introduction for anyone interested in participating in research related to informatics for engineering, but lacking some of the technical skills and background to jump directly into participation on a research project. It is targeted at undergraduates interested in doing research in this area as part of the Informatics Skunkworks program, but open to all that are interested in learning.  The content is open to everyone, it is modular, flexible, and free to use.

This curriculum provides students an introduction to software tools and the associated technical background to understand their use at an introductory level. Throughout the curriculum, students will focus on using two software tools (Citrination and MAST-ML) to generate machine learning models. They will learn key ideas for assessing model performance and decision making skills for how to improve or modify a model.

During the Spring and Fall semesters students are required to attend one synchronous (remote via Zoom) meeting each week as well as periodic community wide activities such as workshops and end of semester meetings. See the Syllabus section for specific semesters to confirm these meeting times. During regular weekly meetings students will be introduced to the core concepts, tools, and activities that are to be completed each week. The weekly activities introduced are then to be completed before the next week’s meeting. The materials are largely based around a set of introductory research modules and professional development activities which are described in more detail in the “Materials” section.

In addition to attending weekly meetings, and completing the weekly activities, the main deliverable that contributes to each students grade (for those receiving course credit) is a weekly slide deck that summarizes the activity completed each week. This set of weekly slide decks, and one Final Slide Deck submitted at the end of the semester contribute to each student’s grade (again for those receiving credit).

It is expected that between synchronous (and asynchronous) meetings, work groups, and completion of activities that students will spend ~9 hours of time on work related to their participation in the course. This typically corresponds to approximately a 2 credit hour course if receiving credit for an independent study course.

Benjamin Afflerbach
Materials Science and Engineering

Specific Syllabi are listed here for recent semesters:

Spring 2022

Fall 2021

The main materials used throughout the course can be found here

For students participating through the University of Wisconsin – Madison, you are able to receive course credit through MSE 299. This course is an independant study course that requires specific permission to enroll. Please reach out to the instructors if you would like to receive credit. Students from other institutions should consult with relevant personnel at their institution about possible credits.  We will provide final materials and suggested grades for all participants.

  1. How much programming, or coding background is required to participate?
    We do not require students to come in with any specific programming experience. There will be significant time spend however running existing python code through Jupyter Notebooks, so some basic familiarity with python may help with understanding the concepts that are being demonstrated by running the existing python code. There will also be activities where students are tasked with making single line replacements to existing code to change settings in the code.