Journal of Complexity, volume 86, pages 101889

High probability bounds on AdaGrad for constrained weakly convex optimization

Publication type: Journal Article
Publication date: 2025-02-01
scimago Q1
wos Q1
SJR: 1.115
CiteScore: 3.1
Impact factor: 1.8
ISSN: 0885064X, 10902708
Abstract
In this paper, we study the high probability convergence of AdaGrad-Norm for constrained, non-smooth, weakly convex optimization with bounded noise and sub-Gaussian noise cases. We also investigate a more general accelerated gradient descent (AGD) template (Ghadimi and Lan, 2016) encompassing the AdaGrad-Norm, the Nesterov's accelerated gradient descent, and the RSAG (Ghadimi and Lan, 2016) with different parameter choices. We provide a high probability convergence rate O˜(1/T) without knowing the information of the weak convexity parameter and the gradient bound to tune the step-sizes.
Found 

Are you a researcher?

Create a profile to get free access to personal recommendations for colleagues and new articles.
Share
Cite this
GOST |
Cite this
GOST Copy
Hong Y. et al. High probability bounds on AdaGrad for constrained weakly convex optimization // Journal of Complexity. 2025. Vol. 86. p. 101889.
GOST all authors (up to 50) Copy
Hong Y., Lin J. High probability bounds on AdaGrad for constrained weakly convex optimization // Journal of Complexity. 2025. Vol. 86. p. 101889.
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1016/j.jco.2024.101889
UR - https://linkinghub.elsevier.com/retrieve/pii/S0885064X24000669
TI - High probability bounds on AdaGrad for constrained weakly convex optimization
T2 - Journal of Complexity
AU - Hong, Yusu
AU - Lin, Junhong
PY - 2025
DA - 2025/02/01
PB - Elsevier
SP - 101889
VL - 86
SN - 0885-064X
SN - 1090-2708
ER -
BibTex
Cite this
BibTex (up to 50 authors) Copy
@article{2025_Hong,
author = {Yusu Hong and Junhong Lin},
title = {High probability bounds on AdaGrad for constrained weakly convex optimization},
journal = {Journal of Complexity},
year = {2025},
volume = {86},
publisher = {Elsevier},
month = {feb},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0885064X24000669},
pages = {101889},
doi = {10.1016/j.jco.2024.101889}
}
Found error?