ResearchTrend.AI
Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems
v1v2v3v4 (latest)

Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems

Stefano Sarao Mannelli
Pierfrancesco Urbani
ArXiv (abs)PDFHTML

Papers citing "Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems"

3 / 3 papers shown
Title
AdaSAM: Boosting Sharpness-Aware Minimization with Adaptive Learning Rate and Momentum for Training Deep Neural Networks
AdaSAM: Boosting Sharpness-Aware Minimization with Adaptive Learning Rate and Momentum for Training Deep Neural Networks
Hao Sun
Li Shen
Qihuang Zhong
Liang Ding
Shi-Yong Chen
Jingwei Sun
Jing Li
Guangzhong Sun
Dacheng Tao
98
34
0
01 Mar 2023
Rigorous dynamical mean field theory for stochastic gradient descent methods
Rigorous dynamical mean field theory for stochastic gradient descent methods
Cédric Gerbelot
Emanuele Troiani
Francesca Mignacco
Florent Krzakala
Lenka Zdeborova
114
29
0
12 Oct 2022
High-dimensional Asymptotics of Langevin Dynamics in Spiked Matrix Models
High-dimensional Asymptotics of Langevin Dynamics in Spiked Matrix Models
Tengyuan Liang
Subhabrata Sen
Pragya Sur
83
7
0
09 Apr 2022