The efficient characterization of quantum states and processes through measurements is central to quantum technologies, be it to certify relevant resources like entanglement, or to extract relevant observables in quantum simulation experiments. The latter deal with complex quantum many-body systems, where the observed data is high-dimensional, making the extraction of quantities of interest notoriously challenging. We discuss possibilities for using modern machine learning methods, which specialize in analyzing large, high-dimensional datasets, to facilitate this task. For a recent review, see arXiv:2509.08011.
