BriefGPT.xyz
May, 2024
一种稳定LPV系统的有限样本泛化界
A finite-sample generalization bound for stable LPV systems
HTML
PDF
Daniel Racz, Martin Gonzalez, Mihaly Petreczky, Andras Benczur, Balint Daroczy
TL;DR
从数据中学习动力系统的一个主要理论挑战是提供关于泛化误差的上界,即在某个有限样本上测量的期望预测误差与经验预测误差之间的差异。本文推导了稳定连续时间线性参数变化系统的一种PAC上界,该上界依赖于所选LPV系统的H2范数,但不依赖于考虑的时间间隔。
Abstract
One of the main theoretical challenges in
learning dynamical systems
from data is providing upper bounds on the
generalization error
, that is, the difference between the expected prediction error and the empirica
→