From differential equation solvers to accelerated first-order methods for convex optimization

被引:0
作者
Hao Luo
Long Chen
机构
[1] Sichuan University,Department of Mathematics
[2] University of California at Irvine,Department of Mathematics
来源
Mathematical Programming | 2022年 / 195卷
关键词
Accelerated first-order methods; Ordinary differential equation; Convergence analysis; Convex optimization; Lyapunov function; Exponential decay; 37N40; 65L20; 65B99; 90C25;
D O I
暂无
中图分类号
学科分类号
摘要
Convergence analysis of accelerated first-order methods for convex optimization problems are developed from the point of view of ordinary differential equation solvers. A new dynamical system, called Nesterov accelerated gradient (NAG) flow, is derived from the connection between acceleration mechanism and A-stability of ODE solvers, and the exponential decay of a tailored Lyapunov function along with the solution trajectory is proved. Numerical discretizations of NAG flow are then considered and convergence rates are established via a discrete Lyapunov function. The proposed differential equation solver approach can not only cover existing accelerated methods, such as FISTA, Güler’s proximal algorithm and Nesterov’s accelerated gradient method, but also produce new algorithms for composite convex optimization that possess accelerated convergence rates. Both the convex and the strongly convex cases are handled in a unified way in our approach.
引用
收藏
页码:735 / 781
页数:46
相关论文
共 43 条
[1]  
Alvarez F(2000)On the minimizing property of a second order dissipative system in Hilbert spaces SIAM J. Control. Optim. 38 1102-1119
[2]  
Attouch H(2000)The heavy ball with friction method, I. The continuous dynamical system: Global exploration of the local minima of a real-valued function by asymptotic analysis of a dissipative dynamical system Commun. Contemp. Math. 2 1-34
[3]  
Goudou X(2016)Fast convergence of inertial dynamics and algorithms with asymptotic vanishing viscosity Math. Program. 168 123-175
[4]  
Redont P(2018)Convergence rates of inertial forward-backward algorithms SIAM J. Optim. 28 849-874
[5]  
Attouch H(2009)A fast iterative shrinkage-thresholding algorithm for linear inverse problems SIAM J. Imag. Sci. 2 183-202
[6]  
Chbani Z(2009)On the long time behavior of second order differential equations with asymptotically small dissipation Trans. Am. Math. Soc. 361 5983-6017
[7]  
Peypouquet J(2007)Asymptotics for some vibro-impact problems with a linear dissipation term J. Math. Pures Appl. 87 291-323
[8]  
Redont P(2009)The gradient and heavy ball with friction dynamical systems: the quasiconvex case Math. Program. 116 173-191
[9]  
Attouch H(1992)New proximal point algorithms for convex minimization SIAM J. Optim. 2 649-664
[10]  
Cabot A(2016)Analysis and Design of Optimization Algorithms via Integral Quadratic Constraints SIAM J. Optim. 26 57-95