1: \begin{abstract}

2: Free-energy-based adaptive biasing methods, such as

3: Metadynamics, the Adaptive Biasing Force (ABF) and their variants, are enhanced sampling algorithms widely used in molecular simulations. Although their efficiency has been empirically acknowledged for decades, providing theoretical insights  via a quantitative convergence analysis is a difficult problem, in particular for the 

4: kinetic Langevin diffusion, which is non-reversible and hypocoercive.

5: We obtain the first exponential convergence result for such a process, in an idealized setting where the dynamics can be associated with a mean-field non-linear flow on the space of probability measures. A key of the analysis is the interpretation of the (idealized) algorithm as the gradient descent of a suitable functional over the space of probability distributions.

6: \end{abstract}

7: