Lorijn Zaadnoordijk

1 Introduction Predictive Processing is a promising unifying account of cognition (Clark, 2013; Friston, 2008, 2010) which has inspired many advances in neuroscience and cognitive science. Predictive Processing builds upon Bayesian theories of the brain and proposes that all cognition is characterized as the comparison between observed input and the expected input based on previous experience, in order to update expectations by incorporating new information.

Abstract The desire to reduce the dependence on curated, labeled datasets and to leverage the vast quantities of unlabeled data has triggered renewed interest in unsupervised (or self-supervised) learning algorithms.