arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.05942cs.LG
收藏

SpReME:多环境动态系统的稀疏回归

SpReME: Sparse Regression for Multi-Environment Dynamic Systems

POSTECH(浦项科技大学)

MoonJeong Park, Youngbin Choi, Namhoon Lee, Dongwoo Kim

摘要:

学习动力系统是科学发现的一个有前途的途径。然而,在多个环境中捕捉支配动态仍然是一个挑战:基于模型的方法依赖于对单个环境的假设的保真度,而基于神经网络的数据驱动方法在推断未来时往往很脆弱。在这项工作中,我们开发了一种称为SpReME的稀疏回归方法,以发现多种环境下的主要动态。具体地说,SpReME在不同的环境中共享一个稀疏的常微分方程(ODE)结构,同时允许每个环境独立地保持ODE项的系数。我们证明,该模型捕捉正确的动态从多个环境中超过四个不同的动态系统,提高了预测性能。

英文摘要:

Learning dynamical systems is a promising avenue for scientific discoveries. However, capturing the governing dynamics in multiple environments still remains a challenge: model-based approaches rely on the fidelity of assumptions made for a single environment, whereas data-driven approaches based on neural networks are often fragile on extrapolating into the future. In this work, we develop a method of sparse regression dubbed SpReME to discover the major dynamics that underlie multiple environments. Specifically, SpReME shares a sparse structure of ordinary differential equation (ODE) across different environments in common while allowing each environment to keep the coefficients of ODE terms independently. We demonstrate that the proposed model captures the correct dynamics from multiple environments over four different dynamic systems with improved prediction performance.

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。

↑