Long-time contractivity estimates for kinetic Kolmogorov-Fokker-Planck equations
Nicolò Forcillo, Alessio Porretta
TL;DR
This work addresses the problem of long-time convergence for hypoelliptic kinetic Kolmogorov-Fokker-Planck equations by introducing a self-contained PDE approach that derives exponential contraction via dual oscillation estimates on the adjoint problem. The authors first prove Wasserstein contraction through a doubling-variables coupling argument and then upgrade to weighted total-variation decay using short-time $L^{\infty}$-to-$W^{1,\infty}$ smoothing (hypocoercivity). The key contributions include (i) a rigorous oscillation-decay framework for the adjoint problem leading to $W_1$-type contraction, (ii) a Bernstein-type short-time smoothing result giving $L^{\infty}$-Lipschitz regularization, and (iii) explicit Lyapunov-function-based conditions yielding exponential decay in the weighted TV norm, with concrete examples such as the Kolmogorov equation. The results provide a robust, duality-based method that extends to nonlinear, inhomogeneous first-order terms and highlights the role of hypoellipticity in driving long-time stabilization without requiring explicit steady states.
Abstract
We prove long-time contractivity estimates and exponential rates of convergence to equilibrium for solutions of hypoelliptic diffusion equations, which include the well-known Kolmogorov equation and similar kinetic Fokker-Planck equations in $\R^d$. Compared to the existing literature, our proof exploits a different approach, elementary and self-contained, based on oscillation estimates for the adjoint problem. We first prove contractivity in Wasserstein distances through doubling variables (coupling) methods. Next, we upgrade the estimate to weighted $L^1$-(or total variation) norms, thanks to short-time hypocoercivity gradient estimates.
