We have coupled adversarial autoencoder deep generative network with adjoined topology optimization technique for metasurface design optimization. We have demonstrated that such approach ensures not only efficient optimization search of highly efficient metasurface designs but also provides unparalleled control over compressed design space distribution. The latter fact assures scalability of the approach to highly-constrained optimization problems.
Paper
Full text
Adversarial Autoencoders for Metasurface Design Optimization (invited)
Semantic Scholar · Engineering · 2020
Abstract
We have coupled adversarial autoencoder deep generative network with adjoined topology optimization technique for metasurface design optimization. We have demonstrated that such approach ensures not only efficient optimization search of highly efficient metasurface designs but also provides unparalleled control over compressed design space distribution. The latter fact assures scalability of the approach to highly-constrained optimization problems.