Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
121 tokens/sec
GPT-4o
9 tokens/sec
Gemini 2.5 Pro Pro
47 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Sublinear Regret with Barzilai-Borwein Step Sizes (2007.14198v3)

Published 28 Jul 2020 in math.OC, cs.SY, and eess.SY

Abstract: This paper considers the online scenario using the Barzilai-Borwein Quasi-Newton Method. In an online optimization problem, an online agent uses a certain algorithm to decide on an objective at each time step after which a possible loss is encountered. Even though the online player will ideally try to make the best decisions possible at each time step, there is a notion of regret associated with the player's decisions. This study examines the regret of an online player using optimization methods like the Quasi-Newton methods, due to their fast convergent properties. The Barzilai-Borwein (BB) gradient method is chosen in this paper over other Quasi-Newton methods such as the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm because of its less computational complexities. In addition, the BB gradient method is suitable for large-scale optimization problems including the online optimization scenario presented in this paper. To corroborate the effectiveness of the Barzilai-Borwein (BB) gradient algorithm, a greedy online gradient algorithm is used in this study based on the two BB step sizes. Through a rigorous regret analysis on the BB step sizes, it is established that the regret obtained from the BB algorithm is sublinear in time. Moreover, this paper shows that the average regret using the online BB algorithm approaches zero without assuming strong convexity on the cost function to be minimized.

Summary

We haven't generated a summary for this paper yet.