A Tutorial on a Lyapunov-Based Approach to the Analysis of Iterative Optimization Algorithms

被引:0
|
作者
Van Scoy, Bryan [1 ]
Lessard, Laurent [2 ]
机构
[1] Miami Univ, Dept Elect & Comp Engn, Oxford, OH 45056 USA
[2] Northeastern Univ, Dept Mech & Ind Engn, Boston, MA 02115 USA
基金
美国国家科学基金会;
关键词
1ST-ORDER; PERFORMANCE; DESIGN;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Iterative gradient-based optimization algorithms are widely used to solve difficult or large-scale optimization problems. There are many algorithms to choose from, such as gradient descent and its accelerated variants such as Polyak's Heavy Ball method or Nesterov's Fast Gradient method. It has long been observed that iterative algorithms can be viewed as dynamical systems, and more recently, as robust controllers. Here, the "uncertainty" in the dynamics is the gradient of the function being optimized. Therefore, worst-case or average-case performance can be analyzed using tools from robust control theory, such as integral quadratic constraints (IQCs). In this tutorial paper, we show how such an analysis can be carried out using an alternative Lyapunov-based approach. This approach recovers the same performance bounds as with IQCs, but with the added benefit of constructing a Lyapunov function.
引用
收藏
页码:3003 / 3008
页数:6
相关论文
共 50 条
  • [1] Lyapunov-Based Stability Analysis of Supervised and Unsupervised Adaptive Algorithms
    de Sousa Junior, Celso
    Attux, Romis
    Suyama, Ricardo
    Romano, Joao M. T.
    2010 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, 2010, : 2988 - 2991
  • [2] Tutorial on Lyapunov-based methods for time-delay systems
    Fridman, Emilia
    EUROPEAN JOURNAL OF CONTROL, 2014, 20 (06) : 271 - 283
  • [3] Lyapunov-based Transient Stability Analysis
    Gao, Jianli
    Chaudhuri, Balarko
    Astolfi, Alessandro
    2022 IEEE 61ST CONFERENCE ON DECISION AND CONTROL (CDC), 2022, : 5099 - 5104
  • [4] Repetitive learning control: A Lyapunov-based approach
    Dixon, WE
    Zergeroglu, E
    Dawson, DM
    Costic, BT
    PROCEEDINGS OF THE 2001 IEEE INTERNATIONAL CONFERENCE ON CONTROL APPLICATIONS (CCA'01), 2001, : 530 - 535
  • [5] Lyapunov-based stability region computation approach
    Schroedel, Frank
    Liu, Hong
    Elghandour, Ramy
    Abel, Dirk
    2015 EUROPEAN CONTROL CONFERENCE (ECC), 2015, : 3009 - 3014
  • [6] Nonlinear analysis and control of a reaction wheel pendulum: Lyapunov-based approach
    Danilo Montoya, Oscar
    Gil-Gonzalez, Walter
    ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH, 2020, 23 (01): : 21 - 29
  • [7] A Lyapunov-based Approach to Safe Reinforcement Learning
    Chow, Yinlam
    Nachum, Ofir
    Duenez-Guzman, Edgar
    Ghavamzadeh, Mohammad
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [8] Repetitive learning control: A Lyapunov-based approach
    Dixon, WE
    Zergeroglu, E
    Dawson, DM
    Costic, BT
    IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2002, 32 (04): : 538 - 545
  • [9] A Lyapunov-Based Methodology for Constrained Optimization with Bandit Feedback
    Cayci, Semih
    Zheng, Yilin
    Eryilmaz, Atilla
    THIRTY-SIXTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FOURTH CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE / THE TWELVETH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2022, : 3716 - 3723
  • [10] Lyapunov-based neural network model predictive control using metaheuristic optimization approach
    Stiti, Chafea
    Benrabah, Mohamed
    Aouaichia, Abdelhadi
    Oubelaid, Adel
    Bajaj, Mohit
    Tuka, Milkias Berhanu
    Kara, Kamel
    SCIENTIFIC REPORTS, 2024, 14 (01):