Work

Online Control with Adversarial Disturbances

Year: 2019

Type: preprint

Abstract: We study the control of a linear dynamical system with adversarial disturbances (as opposed to statistical noise). The objective we consider is one of regret: we desire an online control procedure that can do nearly as well as that of a procedure that has full knowledge of the disturbances in hindsight. Our main result is an efficient algorithm that provides nearly tight regret bounds for this problem. From a technical standpoint, this work generalizes upon previous work in two main aspects: our model allows for adversarial noise in the dynamics, and allows for general convex costs.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.1902.08721

Authors: Agarwal, Naman, Bullins, Brian, Hazan, Elad, Kakade, Sham M., Singh, Karan

Institutions: Google (United States), Princeton University, University of Washington

Language: English

Topic: Advanced Bandit Algorithms Research

Subfield: Management Science and Operations Research

Field: Decision Sciences

Domain: Social Sciences

Open Access status: green

Funders: Washington Research Foundation, Defense Advanced Research Projects Agency, Office of Naval Research

Awards: G4795485331, G8876996369

OpenAlex ID: W2915638893

Cites: 6

Cited by: 124

Related to: 10