The Distinguished Lecture on “Machine Learning for Freeform Structure Design” will take place as follows:
Date: 08 September 2025 (Monday)
Time: 10:30 – 11:30
Venue: Research Building N21, G/F, G013
The speaker is:
Dr. GUO Ruiqi, Postdoctoral Researcher, RIKEN, Japan
The Lecture is:
Machine Learning for Freeform Structure Design
Abstract:
The geometric design of device structures profoundly influences their physical properties and ultimate performance, yet exploring the vast space of freeform geometries and heterogeneous material distributions remains prohibitively time and resource-intensive with conventional numerical simulations. We advocate a machine learning–enabled paradigm for inverse design that treats structures in a non-parameterized, topologically unconstrained form, allowing the deep learning neural networks to capture complex physics directly from data. Once trained, these models function as high-speed, high-accuracy surrogates for mechanics, optics, and electromagnetics, enabling the automated discovery of novel configurations beyond human intuition or template bias. This philosophy unifies design automation across scales and domains—from MEMS resonators and digital composites to optical metasurfaces and radio-frequency circuits—pointing toward a generalized framework where artificial intelligence accelerates the experience-free, data-driven exploration of freeform structures for next-generation device.
Biography:
Dr. GUO Ruiqi is currently a postdoctoral researcher at RIKEN in Japan, working under the supervision of Professor Takao Someya. He earned his Bachelor’s degree in Automotive Engineering from Beihang University in 2018 and completed his Ph.D. in Mechanical Engineering at the University of California, Berkeley, in 2022, under the supervision of Professor Liwei Lin. His primary research focus is on “AI for Science and Engineering.
