Practical and Fast Momentum-Based Power Methods

被引:0
|
作者
Rabbani, Tahseen [1 ]
Jain, Apollo [2 ]
Rajkumar, Arjun [1 ]
Huang, Furong [1 ]
机构
[1] Univ Maryland, Dept Comp Sci, College Pk, MD 20742 USA
[2] Syst & Technol Res, Sensors Div, Arlington, VA USA
基金
美国国家科学基金会;
关键词
matrix decomposition; PCA; power methods; momentum acceleration; streaming PCA;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The power method is a classical algorithm with broad applications in machine learning tasks, including streaming PCA, spectral clustering, and low-rank matrix approximation. The distilled purpose of the vanilla power method is to determine the largest eigenvalue (in absolute modulus) and its eigenvector of a matrix. A momentum-based scheme can be used to accelerate the power method, but achieving an optimal convergence rate with existing algorithms critically relies on additional spectral information that is unavailable at run-time, and sub-optimal initializations can result in divergence. In this paper, we provide a pair of novel momentum-based power methods, which we call the delayed momentum power method (DMPower) and a streaming variant, the delayed momentum streaming method (DMStream). Our methods leverage inexact deflation and are capable of achieving near-optimal convergence with far less restrictive hyperparameter requirements. We provide convergence analyses for both algorithms through the lens of perturbation theory. Further, we experimentally demonstrate that DMPower routinely outperforms the vanilla power method and that both algorithms match the convergence speed of an oracle running existing accelerated methods with perfect spectral knowledge.
引用
收藏
页码:721 / 756
页数:36
相关论文
共 50 条
  • [21] The free retraction of natural rubber: A momentum-based model
    Tunnicliffe, Lewis B.
    Thomas, Alan G.
    Busfield, James J. C.
    POLYMER TESTING, 2015, 47 : 36 - 41
  • [22] On the Global Optimum Convergence of Momentum-based Policy Gradient
    Ding, Yuhao
    Zhang, Junzi
    Lavaei, Javad
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 151, 2022, 151
  • [23] Identification of Human Inertia Properties Using a Momentum-Based Approach
    Lu, Qi
    Ma, Ou
    JOURNAL OF BIOMECHANICAL ENGINEERING-TRANSACTIONS OF THE ASME, 2012, 134 (10):
  • [24] Momentum-based variance-reduced stochastic Bregman proximal gradient methods for nonconvex nonsmooth optimization
    Liao, Shichen
    Liu, Yan
    Han, Congying
    Guo, Tiande
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 266
  • [25] Orbital angular momentum-based Fizeau interferometer measurement system
    Lu, Huali
    Guo, Chenji
    Huang, Xunhua
    Zhao, Hua
    Tang, Wanchun
    OPTICAL DESIGN AND TESTING XII, 2023, 12315
  • [26] Animating reactive motion using momentum-based inverse kinematics
    Komura, T
    Ho, ESL
    Lau, RWH
    COMPUTER ANIMATION AND VIRTUAL WORLDS, 2005, 16 (3-4) : 213 - 223
  • [27] A Momentum-Based Constraint on Optical Flow Tracking of the Endocardial Surface
    Po, Ming Jack
    Lorsakul, Auranuch
    Laine, Andrew F.
    2011 ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC), 2011, : 7191 - 7194
  • [28] Momentum-Based Load Prescriptions: Applications to Jump Squat Training
    Harry, John R.
    Krzyszkowski, John
    Harris, Katie
    Chowning, Luke
    Mackey, Ethan
    Bishop, Chris
    Barker, Leland A.
    JOURNAL OF STRENGTH AND CONDITIONING RESEARCH, 2022, 36 (09) : 2657 - 2662
  • [29] Stability Analysis and Design of Momentum-based Controllers for Humanoid Robots
    Nava, Gabriele
    Romano, Francesco
    Nori, Francesco
    Pucci, Daniele
    2016 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS 2016), 2016, : 680 - 687
  • [30] Momentum-based morphometric analysis with application to Parkinson's disease
    Chen, Jingyun
    Khan, Ali. R.
    McKeown, Martin J.
    Beg, Mirza F.
    MEDICAL IMAGING 2011: VISUALIZATION, IMAGE-GUIDED PROCEDURES, AND MODELING, 2011, 7964