Inverse problems seek to determine the hidden causes that are responsible for observed data. In a PDE-based setting, we assume access to noisy observations of the model solution. Determining the PDE model parameters, initial conditions, or boundary conditions that could have produced the observed data are examples of inverse problems. These problems are often ill-posed and require regularization. After being regularized, inverse problems are typically solved using measurements belonging to a single individual, experiment, or structure. However, in practice, multiple sets of data are often available. For example, in biomedical settings, medical images are taken of not one, but many patients, on a day-to-day basis. Despite its availability, this collection of data measurements, called population data, is not frequently taken into account when solving inverse problems. In this talk, we will look at ways that population data can be included in the resolution of inverse problems, specifically focusing on the construction of population-based regularizations. The theoretical and numerical impact of these population-informed regularizations is discussed for deterministic and Bayesian inverse problems.