Towards Faster Training Algorithms Exploiting Bandit Sampling From Convex to Strongly Convex Conditions-Reference-Cited by-同舟云学术

Towards Faster Training Algorithms Exploiting Bandit Sampling From Convex to Strongly Convex Conditions

Published:2023-04 Issue:2 Volume:7 Page:565-577
ISSN:2471-285X
Container-title:IEEE Transactions on Emerging Topics in Computational Intelligence
language:
Short-container-title:IEEE Trans. Emerg. Top. Comput. Intell.

Author:

Zhou Yangfan¹^ORCID,Huang Kaizhu²^ORCID,Cheng Cheng³,Wang Xuguang³^ORCID,Hussain Amir⁴,Liu Xin¹^ORCID

Affiliation:

1. School of Nano-Tech and Nano-Bionics, University of Science and Technology of China, Hefei City, Anhui Province, China

2. Data Science Research Center, Duke Kunshan University, Kunshan, Jiangsu, China

3. Suzhou Institute of Nano-Tech and Nano-Bionics (SINANO), Chinese Academy of Sciences, Suzhou City, Jiangsu Province, China

4. School of Computing, Edinburgh Napier University, Edinburgh, U.K.

Funder

Suzhou Foreign Experts Project

Chinese Academy of Sciences Key Project

Innovation Workstation of Suzhou Institute of Nano-Tech and Nano-Bionics

National Natural Science Foundation of China

Jiangsu Science and Technology Program

Engineering and Physical Sciences Research Council

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Artificial Intelligence,Computational Mathematics,Control and Optimization,Computer Science Applications

Link

Reference31 articles.

1. Decoupled weight decay regularization;ilya;Proc Int Conf Learn Representations,0

2. Lookahead optimizer: $k$ steps forward, 1 step back;zhang;Proc Adv Neural Inf Process Syst,0

4. Dissecting adam: The sign, magnitude and variance of stochastic gradients;balles;Proc 35th Int Conf Mach Learn,0

5. Lecture 6D - A separate, adaptive learning rate for each connection;hinton;Slides of Lecture Neural Networks for Machine Learning,0