BriefGPT.xyz
Oct, 2023
自适应在线非随机控制
Adaptive Online Non-stochastic Control
HTML
PDF
Naram Mhaisen, George Iosifidis
TL;DR
我们通过设计新的正则化技术,并将其与未经验证的未来成本预测相结合,实现了自适应于环境的Non-stochastic Control算法,这些算法通过考虑系统的内存具有新的数据自适应策略回归界限,并能在准确预测时收缩,即使全部失败时仍保持次线性。
Abstract
We tackle the problem of
non-stochastic control
with the aim of obtaining algorithms that adapt to the controlled environment. Namely, we tailor the
ftrl framework
to
→