Asymptotic behavior of the Arrow-Hurwicz differential system with Tikhonov regularization
Fouad Battahi, Zaki Chbani, Simon K. Niederl\"ander, and Hassan Riahi

TL;DR
This paper studies the long-term behavior of solutions to the Arrow-Hurwicz differential system with Tikhonov regularization, showing convergence to the minimal norm solution and providing decay rate estimates under various conditions.
Contribution
It introduces new conditions on Tikhonov terms that ensure strong convergence and derives explicit decay rates for the solutions and gap functions.
Findings
Solutions converge strongly to the least norm element of zeros.
Decay rates for primal-dual gap and solution velocity are established.
Refined exponential decay estimates are provided for decreasing Tikhonov terms.
Abstract
In a real Hilbert space setting, we investigate the asymptotic behavior of the solutions of the classical Arrow-Hurwicz differential system combined with Tikhonov regularizing terms. Under some newly proposed conditions on the Tikhonov terms involved, we show that the solutions of the regularized Arrow-Hurwicz differential system strongly converge toward the element of least norm within its set of zeros. Moreover, we provide fast asymptotic decay rate estimates for the so-called primal-dual gap function and the norm of the solutions' velocity. If, in addition, the Tikhonov regularizing terms are decreasing, we provide some refined estimates in the sense of an exponentially weighted moving average. Under the additional assumption that the governing operator of the Arrow-Hurwicz differential system satisfies a reverse Lipschitz condition, we further provide a fast rate of strong…
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
TopicsGuidance and Control Systems · Stability and Control of Uncertain Systems · Stability and Controllability of Differential Equations
