Learning in Factored Domains with Information-Constrained Visual Representations

Open in new window