arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2211.01873cs.LGcs.AImath.DSstat.ML
收藏

Port-Metriplectic神经网络:复杂物理系统的图形信息机器学习

Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems

Aragon Institute of Engineering Research (I3A)(阿拉贡工程研究所(I3A));University of Zaragoza(萨拉戈萨大学);Higher Technical School of Industrial Engineering, Polytechnic University of Madrid(马德里理工大学高等工业工程学院);ESI Group(ESI集团);ENSAM Institute of Technology(ENSAM技术学院)

Quercus Hernández, Alberto Badías, Francisco Chinesta, Elías Cueto

摘要:

我们开发的机器学习的复杂的物理系统的基础上端口汉密尔顿形式主义的归纳偏见。为了满足建设的热力学原理在学习物理学(能量守恒,非负熵产生),我们相应地修改端口哈密顿形式主义,以实现端口metriplectic的。我们表明,所构建的网络能够学习复杂系统的物理部分,从而减轻了与这种系统的实验表征和后验学习过程相关的负担。然而,预测可以在整个系统的规模上进行。所提出的技术的性能上的例子。

英文摘要:

We develop inductive biases for the machine learning of complex physical systems based on the port-Hamiltonian formalism. To satisfy by construction the principles of thermodynamics in the learned physics (conservation of energy, non-negative entropy production), we modify accordingly the port-Hamiltonian formalism so as to achieve a port-metriplectic one. We show that the constructed networks are able to learn the physics of complex systems by parts, thus alleviating the burden associated to the experimental characterization and posterior learning process of this kind of systems. Predictions can be done, however, at the scale of the complete system. Examples are shown on the performance of the proposed technique.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑