{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-deep-adversarial-learning-methodology-for","title":"A DEEP ADVERSARIAL LEARNING METHODOLOGY FOR DESIGNING MICROSTRUCTURAL MATERIAL SYSTEMS","arxiv_id":null,"date":"2018-08-26","proceeding":null,"authors":["Xiaolin Li","Zijiang Yang","L. Catherine Brinson","Alok Choudhary","Ankit Agrawal","Wei Chen"],"abstract":"In Computational Materials Design (CMD), it is well recognized that identifying key microstructure characteristics is crucial for determining material design variables. However, existing\r\nmicrostructure characterization and reconstruction (MCR) techniques have limitations to be applied for materials design. Some\r\nMCR approaches are not applicable for material microstructural design because no parameters are available to serve as\r\ndesign variables, while others introduce significant information\r\nloss in either microstructure representation and/or dimensionality reduction. In this work, we present a deep adversarial learning methodology that overcomes the limitations of existing MCR\r\ntechniques. In the proposed methodology, generative adversarial\r\nnetworks (GAN) are trained to learn the mapping between latent\r\nvariables and microstructures. Thereafter, the low-dimensional\r\nlatent variables serve as design variables, and a Bayesian optimization framework is applied to obtain microstructures with desired material property. Due to the special design of the network\r\narchitecture, the proposed methodology is able to identify the latent (design) variables with desired dimensionality, as well as\r\ncapturing complex material microstructural characteristics. The\r\nvalidity of the proposed methodology is tested numerically on a\r\nsynthetic microstructure dataset and its effectiveness for materials design is evaluated through a case study of optimizing optical\r\nperformance for energy absorption. Additional features, such\r\nas scalability and transferability, are also demonstrated in this\r\nwork. In essence, the proposed methodology provides an end-toend solution for microstructural design, in which GAN reduces\r\ninformation loss and preserves more microstructural characteristics, and the GP-Hedge optimization improves the efficiency of\r\ndesign exploration","url_abs":"https://doi.org/10.1115/DETC2018-85633","url_pdf":"http://cucis.ece.northwestern.edu/publications/pdf/LYB18.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-deep-adversarial-learning-methodology-for","repo_url":"https://github.com/abhistar/Microstructure-GAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}