Enhancing Undergraduate Research With Machine Learning
Thank you to everyone that was able to attend the workshop on Oct. 22nd. The discussions in both the Zoom chat as well as the breakout rooms via Zoom was incredibly useful! To help provide easy access to the tools, resources, and content that was covered during the workshop we went back through and tried to pull key things that were referenced or linked to in presentations and compile them in this post. Included in the agenda below are any links or items referenced during the talks as well as a link to the relevant section of the workshop recording.
11:00 AM – Welcome and Introduction to ML and Undergraduate Research (Dane Morgan)
11:25 AM – What you can do: Introduction to the Impact of Machine Learning (ML) in Undergraduate Research (chair – Dane Morgan)
- How Undergraduate Informatics Research Changed my Research, Teaching, and Thinking: A Faculty Perspective (Dane Morgan)
- Working with Undergraduates using Machine Learning Models to Predict Materials Properties: A Graduate Student Perspective (Ben Afflerbach)
- Impact of Machine Learning Focused Research Projects on My Career Path: An Undergraduate perspective (Michael Vanden Heuvel)
11:55 AM – How you can do it: Demonstration of Tools and Resources That Support Undergraduate ML Research (chair – Ben Afflerbach)
- Machine Learning Tools and Resources (Ben Afflerbach)
- Onboarding Materials (Skunkworks Modular Educational Materials, Coursera ML course content, Software Carpentry Programming Modules)
- Cloud computing resources (Google Colab)
- Open Source General ML Software (Scikit-Learn, TensorFlow, PyTorch)
- Research Workflow Software (MAST-ML Documentation, and Github)
- Building a Community (Dane Morgan)
12:15 PM – Identifying and Discussing Community Needs (Breakout Rooms)
- Self-selected breakout rooms themed for researchers and educators
12:35 PM – Breakout Room Report Out and Wrap-up (Dane Morgan, Ben Afflerbach)
- Wrap-up Slide
- Links to followup
12:45 PM – Research Project Brainstorming (Breakout Rooms, current research mentors – optional)
- Self-selected breakout rooms themed based on registration feedback
Full Workshop Recording:
Workshop Planned and Hosted by:
Dane Morgan is the Harvey D. Spangler Professor of Engineering in the Department of Materials Science and Engineering at the University of Wisconsin, Madison. His work combines thermostatistics, thermokinetics, and informatics analysis with atomic scale calculations to understand and predict materials properties. Morgan has graduated/trained over 70 graduate students and postdoctoral researchers and leads the Informatics Skunkworks, which has helped engage over 350 undergraduates at the interface of data science and science and engineering. He has received multiple teaching and research awards, and has published over 350 papers in materials science.
Anne Lynn Gillian-Daniel is the Director of Education and Outreach for the Materials Research Science and Engineering Center (UW-MRSEC) at the University of Wisconsin-Madison. As part of her position, Anne Lynn develops and leads professional development workshops around science communication, mentoring, and bias mitigation for researchers at all stages of their careers. She also works to broaden participation of underrepresented groups in materials science and engineering and to help early career researchers improve their understanding of issues around equity and inclusion.
Rebecca Cors is a social scientist and program evaluator at the Wisconsin Center for Education Research. She has expertise in investigating the effectiveness of science learning programs and has served as an external evaluator for NSF-funded projects designed to train scientists for a stronger workforce. Cors also sat on a review panel for NSF research training (NRT) proposals. She has published and presented research about out-of-school learning, about science and nature education, and about collaborations to promote natural resources management.
Benjamin Afflerbach is a Postdoc in the Department of Materials Science and Engineering at the University of Wisconsin-Madison working with Professor Morgan to help grow the Informatics Skunkworks program. His research focuses on enabling machine learning models to predict materials properties. He has led more than 5 undergraduate research projects and has developed and taught curriculum for onboarding new undergraduate researchers to more than 100 students.