Theorist Overview

AutoRA consists of a set of techniques designed to automate the construction of interpretable models of human brain function and behavior. To approach this problem, we can consider computational models as small, interpretable computation graphs (see also Musslick, 2021). A computation graph can take experiment parameters as input (e.g. the brightness of a visual stimulus) and can transform this input through a combination of functions to produce observable dependent measures as output (e.g. the probability that a participant can detect the stimulus). AutoRA provides adaptations of neural architecture search and symbolic regression to automate the discovery of such models.