Machine Learning and Artificial Intelligence Algorithms in DES Analyses

The size and richness of the Dark Energy Survey (DES) dataset necessitate the development and utilization of automated analysis algorithms to reach the scientific goals of the collaboration. DES scientists employ machine-learning approaches at all stages from data processing to finalized results, and across nearly all subfields of optical survey science: image artifact identification, photometric supernova classification, Milky Way satellite detection, photometric redshift estimation, strong gravitational lensing detection, gravitational wave counterpart identification, galaxy morphology classification, Planet 9 searches, cosmological parameter inference, and several others. In this talk I'll highlight the plethora of algorithms that DES scientists have developed, mention the scientific analyses performed on the DES dataset, and discuss the careful work that has produced robust and novel results with the use of machine learning.