Hessian Riemannian Gradient Flows in Convex Programming

DOI10.1137/S0363012902419977zbMath1077.34050arXiv1811.10331WikidataQ115246809 ScholiaQ115246809MaRDI QIDQ4652545

Felipe Alvarez, Jérôme Bolte, Olivier Brahic

Publication date: 28 February 2005

Published in: SIAM Journal on Control and Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1811.10331

zbMATH Keywords

global convergence; existence; gradient flow; Lyapunov functional; Bregman distance; Legendre transform coordinates; convex and linear programming; Hessian Riemannian metric; Lagrange and Hamilton equations; Legendre-type convex function; quasi-convex minimization

Mathematics Subject Classification ID

90C25: Convex programming

34A12: Initial value problems, existence, uniqueness, continuous dependence and continuation of solutions to ordinary differential equations

34D05: Asymptotic properties of solutions to ordinary differential equations

34C40: Ordinary differential equations and systems on manifolds

Related Items

Gradient Flows, Second-Order Gradient Systems and Convexity, On the Convergence of Gradient-Like Flows with Noisy Gradient Input, Singular Riemannian barrier methods and gradient-projection dynamical systems for constrained optimization, Principal Subspace Flows Via Mechanical Systems on Grassmann Manifolds, An inexact proximal method for quasiconvex minimization, An extension of proximal methods for quasiconvex minimization on the nonnegative orthant, Proximal point method for minimizing quasiconvex locally Lipschitz functions on Hadamard manifolds, Legendre transform and applications to finite and infinite optimization, Central paths in semidefinite programming, generalized proximal-point method and Cauchy trajectories in Riemannian manifolds, Riemannian game dynamics, Stochastic mirror descent dynamics and their convergence in monotone variational inequalities, Learning in games with continuous action sets and unknown payoff functions, Splitting methods with variable metric for Kurdyka-Łojasiewicz functions and general convergence rates, Geometry of matrix decompositions seen through optimal transport and information geometry, Markovian traffic equilibrium, Learning in Games via Reinforcement and Regularization, Full convergence of the proximal point method for quasiconvex functions on Hadamard manifolds, Inertial Game Dynamics and Applications to Constrained Optimization, Penalty-Regulated Dynamics and Robust Learning Procedures in Games, Unnamed Item