arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2002.08071cs.LGcs.NEstat.ML
收藏

剖析神经常微分方程

Dissecting Neural ODEs

The University of Tokyo(东京大学);KAIST(韩国科学技术院)

Stefano Massaroli, Michael Poli, Jinkyoo Park, Atsushi Yamashita, Hajime Asama

摘要:

连续深度学习架构最近以神经常微分方程(Neural ODE)的形式重新出现。这种无限深度的方法在理论上弥合了深度学习和动力系统之间的差距,提供了一个新的视角。然而,破译这些模型的内部工作仍然是一个开放的挑战,因为大多数应用程序将它们作为通用的黑盒模块。在这项工作中,我们“打开盒子”,进一步发展的连续深度配方,目的是澄清几个设计选择对底层动力学的影响。

英文摘要:

Continuous deep learning architectures have recently re-emerged as Neural Ordinary Differential Equations (Neural ODEs). This infinite-depth approach theoretically bridges the gap between deep learning and dynamical systems, offering a novel perspective. However, deciphering the inner working of these models is still an open challenge, as most applications apply them as generic black-box modules. In this work we "open the box", further developing the continuous-depth formulation with the aim of clarifying the influence of several design choices on the underlying dynamics.

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。

↑