Free energy computations by minimization of Kullback-Leibler divergence: An efficient adaptive biasing potential method for sparse representations (Q417933)

The present paper proposes an adaptive biasing potential technique for the computation of free energy landscapes, providing a unifying framework for estimating the free energy function simultaneously with biasing the dynamics. The minimization of the Kullback-Leibler divergence in the extended space provides rigorous convergence bounds and diagnostics. It employs parallelizable sampling schemes that are based on the adaptive sequential Monte Carlo method. The sequential nature of the learning and sampling scheme enables the efficient calculation of free energy functions parametrized by the temperature. The characteristics and capabilities of the proposed method are demonstrated on three numerical examples.