Abstract:In the framework of real Hilbert spaces, we investigate first-order dynamical systems governed by monotone and continuous operators. We demonstrate that when the monotone operator flow is augmented with a Tikhonov regularization term, the resulting trajectory converges strongly to the element of the set of zeros with minimal norm. In addition, rates of convergence in norm for the trajectory's velocity and the operator along the trajectory can be derived in terms of the regularization function. In some particular cases, these rates of convergence can outperform the ones of the coercive operator flows and can be as fast as $O(\frac{1}{t})$ as $t \rightarrow +\infty$. In this way, we emphasize a surprising acceleration feature of the Tikhonov regularization. Additionally, we explore these properties for monotone operator flows that incorporate time rescaling and an anchor point and show that they are closely linked to second-order dynamics with a vanishing damping term. The convergence and convergence rate results we achieve for these systems complement recent findings for the Fast Optimistic Gradient Descent Ascent (OGDA) dynamics. When the monotone operator is defined as the identity minus a nonexpansive operator, the monotone equations transform into a fixed point problem. In such cases, explicitly discretizing the system with Tikhonov regularization, enhanced by an anchor point, leads to the Halpern fixed point iteration. We identify two regimes for the regularization sequence which ensure that the generated sequence of iterates converges strongly to the fixed point nearest to the anchor point. Furthermore, we establish a general theoretical framework that provides convergence rates for the vanishing of the discrete velocity and the fixed point residual. For certain regularization sequences, we derive specific convergence rates that align with those observed in continuous time.

Trajectory Convergence from Coordinate-wise Decrease of General Energy Functions

Accelerated Gradient Dynamics on Riemannian Manifolds: Faster Rate and Trajectory Convergence

On the convergence of trajectory statistical solutions

Strong convergence and fast rates for systems with Tikhonov regularization

Costate Convergence with Legendre-Lobatto Collocation for Trajectory Optimization

Inertial dynamics with vanishing Tikhonov regularization for multobjective optimization

Convergence Properties of Newton's Method for Globally Optimal Free Flight Trajectory Optimization

Exploring exact-factorization-based trajectories for low-energy dynamics near a conical intersection

Fast convergence rates and trajectory convergence of a Tikhonov regularized inertial primal\mbox{-}dual dynamical system with time scaling and vanishing damping

Strong convergence towards the minimum norm solution via temporal scaling and Tikhonov approximation of a first-order dynamical system

Data-driven Evolutions of Critical Points

Constrained similarity of 2-D trajectories by minimizing the H1 semi-norm of the trajectory difference

Almost sure convergence of stochastic Hamiltonian descent methods

Asymptotics for A Dissipative Dynamical System with Linear and Gradient-Driven Damping

Trajectory-Oriented Control Using Gradient Descent: An Unconventional Approach

Tikhonov regularization of monotone operator flows not only ensures strong convergence of the trajectories but also speeds up the vanishing of the residuals

A Bell-Evans-Polanyi principle for molecular dynamics trajectories and its implications for global optimization

Ergodic Trajectory Optimization on Generalized Domains Using Maximum Mean Discrepancy

Trajectory Consensus for Coordination of Multiple Curvature-Bounded Vehicles

Solving convex optimization problems via a second order dynamical system with implicit Hessian damping and Tikhonov regularization

Minimal Curvature Trajectories: Riemannian Geometry Concepts for Model Reduction in Chemical Kinetics