On time-splitting methods for gradient flows with two dissipation mechanisms
Alexander Mielke, Riccarda Rossi, Artur Stephan

TL;DR
This paper introduces a split-step method for solving gradient flows with two dissipation mechanisms, allowing solutions to be constructed by alternating between simpler gradient systems, with convergence guaranteed by energy-dissipation principles.
Contribution
It develops a novel split-step approach for gradient flows with dual dissipation potentials, enabling solution construction via semiflows and minimizing movements with proven convergence.
Findings
Solutions can be constructed by alternating simpler gradient flows.
Convergence is established using energy-dissipation principles.
Method applies to Banach space gradient systems with combined dissipation.
Abstract
We consider generalized gradient systems in Banach spaces whose evolutions are generated by the interplay between an energy functional and a dissipation potential. We focus on the case in which the dual dissipation potential is given by a sum of two functionals and show that solutions of the associated gradient-flow evolution equation with combined dissipation can be constructed by a split-step method, i.e. by solving alternately the gradient systems featuring only one of the dissipation potentials and concatenating the corresponding trajectories. Thereby the construction of solutions is provided either by semiflows, on the time-continuous level, or by using Alternating Minimizing Movements in the time-discrete setting. In both cases the convergence analysis relies on the energy-dissipation principle for gradient systems.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Mathematical Modeling in Engineering · Stability and Controllability of Differential Equations · Mathematical Biology Tumor Growth
