Unsupervised Progressive Learning and the STAM Architecture

Abstract

We first pose the Unsupervised Progressive Learning (UPL) problem: an online representation learning problem in which the learner observes a non-stationary and unlabeled data stream, and identifies a growing number of features that persist over time even though the data is not stored or replayed. To solve the UPL problem we propose the Self-Taught Associative Memory (STAM) architecture. Layered hierarchies of STAM modules learn based on a combination of online clustering, novelty detection, forgetting outliers, and storing only prototypical features rather than specific examples. We evaluate STAM representations using classification and clustering tasks. Even though there are no prior approaches that are directly applicable to the UPL problem, we evaluate the STAM architecture in comparison to some unsupervised and self-supervised deep learning approaches adapted in the UPL context.