Mini-batch Gradient Descent
Mini-Batch Gradient Descent is a gradient-based optimization algorithm that, as a variant of stochastic gradient descent, computes parameter updates using the average gradient of a small, randomly sampled subset of training data—termed a mini-batch—thereby balancing the computational efficiency of stochastic updates with the stability of batch gradient descent.
197
Publications
14.3K
Citations
599
Authors
242
Institutions
Publications per year
2017–2026
148
Variants
Mini-batch Sgd
Parents
197
Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
Priya Goyal, Piotr Dollár, Ross Girshick et al. · arXiv (Cornell University) · 2017 · 2.6K citations · Full text
Improved Deep Embedded Clustering with Local Structure Preservation
Xifeng Guo, Long Gao, Xinwang Liu et al. · 2017 · 900 citations · Full text
Learning to Reweight Examples for Robust Deep Learning
Mengye Ren, Wenyuan Zeng, Bin Yang et al. · arXiv (Cornell University) · 2018 · 581 citations · Full text
Optimal Distributed Online Prediction using Mini-Batches
Ofer Dekel, Ran Gilad-Bachrach, Ohad Shamir et al. · arXiv (Cornell University) · 2010 · 563 citations · Full text
1–5 of 197
599
7
h-index
University of North Carolina at Chapel Hill, United States
5 publications · 180 citations
5
h-index
4
h-index
4
h-index
4
h-index
1–5 of 599
242

University of Edinburgh
Edinburgh, United Kingdom
7 h-index
10 publications
1.1K citations
Microsoft (United States)
Redmond, United States
6 h-index
12 publications
2.8K citations

Johns Hopkins University
Baltimore, United States
5 h-index
9 publications
301 citations

Lehigh University
Bethlehem, United States
5 h-index
6 publications
801 citations
Dalian University of Technology
Dalian, China
5 h-index
12 publications
618 citations
Google (United States)
Mountain View, United States
4 h-index
13 publications
615 citations

Sun Yat-sen University
Guangzhou, China
4 h-index
8 publications
123 citations
Toyota Technological Institute at Chicago
Chicago, United States
4 h-index
7 publications
736 citations
1–8 of 242
Repository · 63 publications · 7.1K citations
29
h-index
8 publications · 185 citations
7
h-index
5 publications · 577 citations
5
h-index
5 publications · 210 citations
5
h-index
TFS · 4 publications · 374 citations
4
h-index
1–5