Geometry Encoding for Numerical Simulations
Abstract
We present a notion of geometry encoding suitable for machine learning-based numerical simulation. In particular, we delineate how this notion of encoding is different than other encoding algorithms commonly used in other disciplines such as computer vision and computer graphics. We also present a model comprised of multiple neural networks including a processor, a compressor and an evaluator. These parts each satisfy a particular requirement of our encoding. We compare our encoding model with the analogous models in the literature.
Cite
Text
Maleki et al. "Geometry Encoding for Numerical Simulations." ICLR 2021 Workshops: GTRL, 2021.Markdown
[Maleki et al. "Geometry Encoding for Numerical Simulations." ICLR 2021 Workshops: GTRL, 2021.](https://mlanthology.org/iclrw/2021/maleki2021iclrw-geometry/)BibTeX
@inproceedings{maleki2021iclrw-geometry,
title = {{Geometry Encoding for Numerical Simulations}},
author = {Maleki, Amir and Heyse, Jan and Ranade, Rishikesh and He, Haiyang and Kasimbeg, Priya and Pathak, Jay},
booktitle = {ICLR 2021 Workshops: GTRL},
year = {2021},
url = {https://mlanthology.org/iclrw/2021/maleki2021iclrw-geometry/}
}