The unprecedented combination of depth and area will enable LSST to map the evolution of galaxies in exquisite detail over cosmic time. The sheer volume and complexity of the data will demand novel analysis techniques, in particular those that offer high levels of automation. A growing convergence between astrophysics and computer science is revealing the potential of artificial intelligence and machine learning techniques in the rapid exploitation and analysis of future surveys like LSST. Examples of these are described in the AI and Machine Learning page of this working group.
While novel techniques will enable us to extract and analyse information from LSST data, our understanding of the physics of galaxy evolution relies on the comparison of this data with theoretical models, in particular simulations in cosmological volumes. These comparisons are likely to combine suites of semi-analytical models and hydro-dynamical simulations in which baryons and dark matter are evolved self-consistently. While the former lend themselves well to the rapid exploration of parameter space, the latter offer detailed three-dimensional predictions of the internal properties of the baryons, which can be strongly constrained by the high-resolution data from instruments like LSST. Examples of predictions from state-of-the-art cosmological simulations are described in the Cosmological Simulations page of this working group.