Session: DAC-04-01-Data-Driven Design

Paper Number: 87653

87653 - t-METASET: Task-Aware Generation of Metamaterial Datasets by Diversity-Based Active Learning 

Inspired by the recent success of deep learning in diverse domains, data-driven metamaterials design has emerged as a compelling design paradigm to unlock the potential of multiscale architecture. However, existing model-centric approach lacks principled methodologies dedicated to high-quality data generation. Resorting to naïve sampling, existing metamaterial datasets suffer from property distributions that are either highly imbalanced or at odds with design tasks of interest. To this end, we propose t-METASET: an intelligent data acquisition framework that aims at task-aware dataset generation. We seek a solution to a commonplace yet overlooked scenario at early design stages: when a massive shape library has been prepared with no properties evaluated. The key idea is to exploit a data-driven shape descriptor learned from generative models, fit a sparse regressor as the start-up agent, and leverage diversity-related metrics to drive data acquisition to areas that help designers fulfill design goals. We validate the proposed framework in three hypothetical deployment scenarios, which encompass general use, task-aware use, and tailorable use, using two large-scale shape-only mechanical metamaterial datasets. The results demonstrate that t-METASET can incrementally grow task-aware datasets. Applicable to general design representations, t-METASET can boost future advancements of not only metamaterials but data-driven design in other domains.         

Presenting Author: Doksoo Lee Northwestern University, Dept. of Mechanical Engineering

Presenting Author Biography: Doksoo Lee is a PhD candidate at Northwestern University (advisor: Prof. Wei Chen). His current research encompasses data-driven multiscale design; machine learning; data quality management; topology optimization; Gaussian processes; Bayesian optimization; mechanical/optical metamaterials.<br/><br/>Doksoo Lee received his B.S. and M.S. in mechanical engineering Yonsei University in 2017 and 2019 (advisor: Prof. Jeonghoon Yoo), respectively. He was a summer intern of Department of Materials Durability CAE, MDO, and AI Methods at Ford Motors Company in 2021.

Authors:

Doksoo Lee Northwestern University, Dept. of Mechanical Engineering
Yu-Chin Chan Siemens Corporate Technology
Wei (Wayne) Chen Northwestern University, Dept. of Mechanical Engineering
Liwei Wang Shanghai Jiao Tong University, School of Mechanical Engineering
Wei Chen Northwestern University, Dept. of Mechanical Engineering