A machine learning inversion scheme for determining interaction from scattering

Abstract Small angle scattering techniques have now been routinely used to quantitatively determine the potential of mean force in colloidal suspensions. However numerical accuracy data interpretation is often compounded by approximations adopted liquid state analytical theories. To circumvent this long standing issue, here we outline a machine learning strategy for determining effective interaction condensed phases matter using scattering. Via case study suspensions, show that can be probabilistically inferred from spectra without any restriction imposed model assumptions. Comparisons existing parametric approaches demonstrate superior performance method accuracy, efficiency, and applicability. This effectively enable quantification highly correlated systems diffraction experiments.