BriefGPT.xyz
Feb, 2020
高效学习线性二次调节器的对数损失
Logarithmic Regret for Learning Linear Quadratic Regulators Efficiently
HTML
PDF
Asaf Cassel, Alon Cohen, Tomer Koren
TL;DR
本文介绍了Linear Quadratic Control系统的学习问题和非常高效的算法，算法的遗憾只随着决策步数的对数级别增加，并且当某些特定条件成立时可以得到更好的结果，但当条件不成立时，无法避免遗憾增长的平方根级别。
Abstract
We consider the problem of learning in
linear quadratic control systems
whose transition parameters are initially unknown. Recent results in this setting have demonstrated efficient
learning algorithms
with
→