ParaFlow: Parareal Acceleration of Gradient-Flow Minimization
Keywords
Advanced Numerical Methods for Scientific Computing
Code:
51/2026
Title:
ParaFlow: Parareal Acceleration of Gradient-Flow Minimization
Date:
Tuesday 23rd June 2026
Author(s):
Bellezza P.; Ciaramella G.; Macchini C.; Mazzieri I.; Verani M.
Download link:
Abstract:
This work presents the ParaFlow class of optimization algorithms and its specific realization, the ParaFlowS algorithm. The ParaFlow framework employs the Parareal algorithm to enhance the convergence rate of gradient flows towards a minimum. The ParaFlowS method integrates the Parareal approach with (potentially stochastic) gradient descent (GD) method, resulting in a purely sequential optimization strategy. The proposed acceleration framework is assessed through extensive numerical experiments on unconstrained optimization problems associated with the training of fully connected neural networks.
This report, or a modified version of it, has been also submitted to, or published on
Proceeding of the 29th International Conference on Domain Decomposition Methods.
Proceeding of the 29th International Conference on Domain Decomposition Methods.