Revisiting Smoothed Online Learning

by   Moshe Y. Vardi, et al.

In this paper, we revisit the problem of smoothed online learning, in which the online learner suffers both a hitting cost and a switching cost, and target two performance metrics: competitive ratio and dynamic regret with switching cost. To bound the competitive ratio, we assume the hitting cost is known to the learner in each round, and investigate the greedy algorithm which simply minimizes the weighted sum of the hitting cost and the switching cost. Our theoretical analysis shows that the greedy algorithm, although straightforward, is 1+ 2/α-competitive for α-polyhedral functions, 1+O(1/λ)-competitive for λ-quadratic growth functions, and 1 + 2/√(λ)-competitive for convex and λ-quadratic growth functions. To bound the dynamic regret with switching cost, we follow the standard setting of online convex optimization, in which the hitting cost is convex but hidden from the learner before making predictions. We modify Ader, an existing algorithm designed for dynamic regret, slightly to take into account the switching cost when measuring the performance. The proposed algorithm, named as Smoothed Ader, attains an optimal O(√(T(1+P_T))) bound for dynamic regret with switching cost, where P_T is the path-length of the comparator sequence. Furthermore, if the hitting cost is accessible in the beginning of each round, we obtain a similar guarantee without the bounded gradient condition.


page 1

page 2

page 3

page 4


Understand Dynamic Regret with Switching Cost for Online Decision Making

As a metric to measure the performance of an online method, dynamic regr...

Smoothed Online Convex Optimization Based on Discounted-Normal-Predictor

In this paper, we investigate an online prediction strategy named as Dis...

Optimal Parameter-free Online Learning with Switching Cost

Parameter-freeness in online learning refers to the adaptivity of an alg...

Movement Penalized Bayesian Optimization with Application to Wind Energy Systems

Contextual Bayesian optimization (CBO) is a powerful framework for seque...

Competitive ratio versus regret minimization: achieving the best of both worlds

We consider online algorithms under both the competitive ratio criteria ...

Smoothed Online Optimization for Regression and Control

We consider Online Convex Optimization (OCO) in the setting where the co...

Online Optimization with Feedback Delay and Nonlinear Switching Cost

We study a variant of online optimization in which the learner receives ...

Please sign up or login with your details

Forgot password? Click here to reset