Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently

Haoyuan Sun, Kwangjun Ahn, Christos Thrampoulidis, Navid Azizan

Type: Preprint

Publication Date: 2022-01-01

Citations: 1

DOI: https://doi.org/10.48550/arxiv.2205.12808

View Publication

Locations

arXiv (Cornell University) - View - PDF
DataCite API - View

Similar Works

Action	Title	Year	Authors
+	A Unified Approach to Controlling Implicit Regularization via Mirror Descent	2023	Haoyuan Sun Khashayar Gatmiry Kwangjun Ahn Navid Azizan
+	Stochastic Mirror Descent on Overparameterized Nonlinear Models: Convergence, Implicit Regularization, and Generalization	2019	Navid Azizan Sahin Lale Babak Hassibi
+	Achieving Margin Maximization Exponentially Fast via Progressive Norm Rescaling	2023	Mingze Wang Zeping Min Lei Wu
+ PDF Chat	A Survey of Optimization Methods for Training DL Models: Theoretical Perspective on Convergence and Generalization	2025	Jing Wang Anna Choromanska
+ PDF Chat	Optimizing Attention with Mirror Descent: Generalized Max-Margin Token Selection	2024	Aaron Alvarado Kristanto Julistiono Davoud Ataee Tarzanagh Navid Azizan
+	Gradient Descent Maximizes the Margin of Homogeneous Neural Networks	2019	Kaifeng Lyu Jian Li
+	Large Margin Deep Neural Networks: Theory and Algorithms.	2015	Shizhao Sun Wei Chen Liwei Wang Tie‐Yan Liu
+	Penalizing Gradient Norm for Efficiently Improving Generalization in Deep Learning	2022	Yang Zhao Hao Zhang Xiuyuan Hu
+	Gradient Descent Finds Global Minima for Generalizable Deep Neural Networks of Practical Sizes	2019	Kenji Kawaguchi Jiaoyang Huang
+	Explicit Regularization via Regularizer Mirror Descent	2022	Navid Azizan Sahin Lale Babak Hassibi
+	Towards understanding how momentum improves generalization in deep learning	2022	Samy Jelassi Yuanzhi Li
+	On the Maximum Hessian Eigenvalue and Generalization	2022	Simran Kaur Jérémy Cohen Zachary C. Lipton
+	Exponential convergence rates for Batch Normalization: The power of length-direction decoupling in non-convex optimization	2018	Jonas Köhler Hadi Daneshmand Aurélien Lucchi Ming Zhou Klaus Neymeyr Thomas Hofmann
+	Exponential convergence rates for Batch Normalization: The power of length-direction decoupling in non-convex optimization	2018	Jonas Köhler Hadi Daneshmand Aurélien Lucchi Ming Zhou Klaus Neymeyr Thomas Hofmann
+	Recent Advances in Large Margin Learning	2021	Yiwen Guo Changshui Zhang
+ PDF Chat	Recent Advances in Large Margin Learning	2021	Yiwen Guo Changshui Zhang
+ PDF Chat	Recent Advances in Large Margin Learning	2021	Yiwen Guo Changshui Zhang
+	Gradient Descent Provably Optimizes Over-parameterized Neural Networks	2018	Simon S. Du Xiyu Zhai Barnabás Póczos Aarti Singh
+	Gradient Descent Provably Optimizes Over-parameterized Neural Networks	2018	Simon S. Du Xiyu Zhai Barnabás Póczos Aarti Singh
+	Gradient Descent Provably Optimizes Over-parameterized Neural Networks	2018	Simon S. Du Xiyu Zhai Barnabás Póczos Aarti Singh

Works That Cite This (0)

Action	Title	Year	Authors

Works Cited by This (0)

Action	Title	Year	Authors