BriefGPT.xyz
Jun, 2018
在线线性二次控制
Online Linear Quadratic Control
HTML
PDF
Alon Cohen, Avinatan Hassidim, Tomer Koren, Nevena Lazic, Yishay Mansour...
TL;DR
我们研究了控制具有已知嘈杂动力学和对抗选择二次损失的线性时不变系统的问题,并提出了第一种在这种情况下保证O(sqrt(T))遗憾的有效在线学习算法。我们的算法依赖于对系统稳态分布的新型SDP松弛。与以前提出的松弛相反,我们的SDP的可行解都对应于“强稳定”策略,这些策略混合到稳定状态的速度呈指数增长。
Abstract
We study the problem of controlling
linear time-invariant systems
with known noisy dynamics and adversarially chosen quadratic losses. We present the first efficient
online learning algorithms
in this setting tha
→