# Dynamic Regret of Convex and Smooth Functions

7 Jul 2020Peng ZhaoYu-Jie ZhangLijun ZhangZhi-Hua Zhou

We investigate online convex optimization in non-stationary environments and choose the dynamic regret as the performance measure, defined as the difference between cumulative loss incurred by the online algorithm and that of any feasible comparator sequence. Let $T$ be the time horizon and $P_T$ be the path-length that essentially reflects the non-stationarity of environments, the state-of-the-art dynamic regret is $\mathcal{O}(\sqrt{T(1+P_T)})$... (read more)

PDF Abstract

No code implementations yet. Submit your code now