Tamas Budavari, Affiliation: Johns Hopkins University Bayesian Inference from Photometric Surveys With the upcoming survey telescopes just around the corner, the statistical and computational challenges of astronomy are more prominent than ever. I will discuss some of the fundamental issues that are at the core of our photometric analyses. A Bayesian approach is introduced for cross-identifying astronomical sources, which is extendable, e.g., to incorporate models of spectral energy distributions, or to accommodate the proper motion of stars. Probabilistic inferences open up new possibilities for determining properties of celestial objects based on their photometric observations. Constraints on photometric redshifts and other physical parameters in the more general inversion problem are derived from first principles that also point us toward the next steps.