Presentation | 2022-03-09 [Invited Talk] --- Masanobu Horie, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Learning flow phenomena is an important problem for both theoretical and practical aspects. Graph neural networks are a promising approach to learning physical simulation, including computational fluid dynamics. However, due to the data-driven nature of machine learning, conventional GNN’s prediction can be unstable due to the complexity of 3D shapes and physical states. This research presents a graph neural network model that intrinsically equips isometric transformation equivariance, which is required to consider physical symmetries regarding rotation, translation, and reflection. We demonstrate that the proposed model can stably predict fluid phenomena thanks to its equivariance. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Graph Neural Network / Equivariance / Physical Simulation / Fluid Dynamics |
Paper # | IBISML2021-43 |
Date of Issue | 2022-03-01 (IBISML) |
Conference Information | |
Committee | IBISML |
---|---|
Conference Date | 2022/3/8(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Machine Learning, etc. |
Chair | Ichiro Takeuchi(Nagoya Inst. of Tech.) |
Vice Chair | Masashi Sugiyama(Univ. of Tokyo) |
Secretary | Masashi Sugiyama(Univ. of Tokyo) |
Assistant | Tomoharu Iwata(NTT) / Atsuyoshi Nakamura(Hokkaido Univ.) |
Paper Information | |
Registration To | Technical Committee on Infomation-Based Induction Sciences and Machine Learning |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Invited Talk] --- |
Sub Title (in English) | |
Keyword(1) | Graph Neural Network |
Keyword(2) | Equivariance |
Keyword(3) | Physical Simulation |
Keyword(4) | Fluid Dynamics |
1st Author's Name | Masanobu Horie |
1st Author's Affiliation | Research Institute for Computational Science Co. Ltd.(RICOS Co. Ltd.) |
Date | 2022-03-09 |
Paper # | IBISML2021-43 |
Volume (vol) | vol.121 |
Number (no) | IBISML-419 |
Page | pp.pp.37-37(IBISML), |
#Pages | 1 |
Date of Issue | 2022-03-01 (IBISML) |