OVER-PARAMETERIZED MODEL OPTIMIZATION WITH POLYAK-LOJASIEWICZ CONDITION

被引：0
|
作者：
Chen, Yixuan [1 ]
Shi, Yubin [1 ]
Dong, Mingzhi [1 ]
Yang, Xiaochen [2 ]
Li, Dongsheng [3 ]
Wang, Yujiang [4 ]
Dick, Robert P. [5 ]
Lv, Qin [6 ]
Zhao, Yingying [1 ]
Yang, Fan [7 ]
Gu, Ning [1 ]
Shang, Li [1 ]
机构：
[1] China and Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University, Shanghai, China
[2] School of Mathematics Statistics, The University of Glasgow, Glasgow, United Kingdom
[3] Microsoft Research Asia, Shanghai, China
[4] Department of Engineering Science, University of Oxford, Oxford, United Kingdom
[5] Department of Electrical Engineering and Computer Science, University of Michigan, Michigan, United States
[6] Department of Computer Science, University of Colorado Boulder, Boulder,CO, United States
[7] School of Microelectronics, Fudan University, Shanghai, China
来源：
11th International Conference on Learning Representations, ICLR 2023 | 2023年
关键词：
Compilation and indexing terms; Copyright 2025 Elsevier Inc;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
Efficiency - Number theory - Parameterization
引用
收藏
相关论文
共 50 条
  • [41] Convex Geometry and Duality of Over-parameterized Neural Networks
    Ergen, Tolga
    Pilanci, Mert
    JOURNAL OF MACHINE LEARNING RESEARCH, 2021, 22
  • [42] Implicit Regularization in Over-Parameterized Support Vector Machine
    Sui, Yang
    He, Xin
    Bai, Yang
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [43] Over-Parameterized Optical Flow Using a Stereoscopic Constraint
    Rosman, Guy
    Shem-Tov, Shachar
    Bitton, David
    Nir, Tal
    Adiv, Gilad
    Kimmel, Ron
    Feuer, Arie
    Bruckstein, Alfred M.
    SCALE SPACE AND VARIATIONAL METHODS IN COMPUTER VISION, 2012, 6667 : 761 - +
  • [44] On the Computational and Statistical Complexity of Over-parameterized Matrix Sensing
    Zhuo, Jiacheng
    Kwon, Jeongyeol
    Ho, Nhat
    Caramanis, Constantine
    JOURNAL OF MACHINE LEARNING RESEARCH, 2024, 25 : 1 - 47
  • [45] Convex geometry and duality of over-parameterized neural networks
    Ergen, Tolga
    Pilanci, Mert
    Journal of Machine Learning Research, 2021, 22
  • [46] Global Convergence of Over-parameterized Deep Equilibrium Models
    Ling, Zenan
    Xie, Xingyu
    Wang, Qiuhao
    Zhang, Zongpeng
    Lin, Zhouchen
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 206, 2023, 206 : 767 - 787
  • [47] Further remarks on constrained over-parameterized linear models
    Nesrin Güler
    Melek Eriş Büyükkaya
    Statistical Papers, 2024, 65 : 975 - 988
  • [48] Further remarks on constrained over-parameterized linear models
    Guler, Nesrin
    Buyukkaya, Melek Eris
    STATISTICAL PAPERS, 2024, 65 (02) : 975 - 988
  • [49] Provable Acceleration of Heavy Ball beyond Quadratics for a Class of Polyak-Lojasiewicz Functions when the Non-Convexity is Averaged-Out
    Wang, Jun-Kun
    Lin, Chi-Heng
    Wibisono, Andre
    Hu, Bin
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022,
  • [50] Convergence beyond the over-parameterized regime using Rayleigh quotients
    Robin, David A. R.
    Scaman, Kevin
    Lelarge, Marc
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35, NEURIPS 2022, 2022,
12345