Date: Wednesday, February 28th, 2:00p.m.
Location: Engineering and Physical Sciences Library, Math (Kirwan) Building
The nature of product design has increased in scale, both inside corporations and in self-organized online communities (e.g., OpenIDEO, Local Motors). This is thanks to unprecedented amounts of digital design information made possible by globally distributed groups of thousands of people who collaborate together on design projects over the Internet. However, this increased scale and diversity comes with a price: 1) these groups generate more data than they can effectively use, 2) it becomes difficult to leverage their diverse expertise, and 3) involving non-experts meaningfully in the design process, particularly for complex mechanical systems, requires rethinking how people interact with design tools and what kind of intelligent support we need to provide.
In this talk I'll address how advances in machine learning can ameliorate these issues. Specifically, my students and I will introduce ongoing work on three problems: 1) how to use data to understand and simplify complex, high-dimensional, design spaces (to aid in techniques like optimization, design synthesis, and design exploration), 2) how to filter high-quality, diverse submissions out of large pools of design ideas generated by online communities (to aid in design generation and selection), and 3) how to enable non-experts to design complex mechanical parts (such as 3D printable robots) by using AI to automate various mechanical design tasks. Each problem highlights how building probabilistic models of designs via data can often produce a whole that is greater than the sum of its parts and make design (even of complex, physical systems) more inclusive.
Mark Fuge is an Assistant Professor of Mechanical Engineering at the University of Maryland, College Park, and recently joined the HCIL as a faculty member. His research lies at the intersection of Mechanical Engineering, Machine Learning, and Design; an area often referred to as "Design Informatics" or "Data-Driven Design." He received his Ph.D. at the University of California at Berkeley, and received his M.S. and B.S. at Carnegie Mellon University. He has conducted research in applied machine learning, optimization, network analysis, additive manufacturing, human-computer interfaces, crowdsourcing, and creativity support tools. He has received a DARPA Young Faculty Award and a National Defense Science and Engineering Graduate (NDSEG) Fellowship. You can learn more about his research at his lab’s website: https://ideal.umd.edu