Learning to Theorize the World from Observation

URL

Notes

Abstract

Contemporary world models often operationalize understanding as accurate future prediction in latent or observation space ()

human understanding emerges through the construction of internal theories of how the world works ()

Learning-to-Theorize, a learning paradigm for inferring explicit explanatory theories of the world from raw, non-textual observations ()

Neural Theorizer (NEO), a probabilistic neural model that induces latent programs as a learned Language of Thought ()