Theory/foundations  (e.g. sample size bounds for maximum likelihood estimation,  dictionary learning, and boolean function learning); complexity theory; geometric modeling and constraint solving; algorithms and discrete modeling.