Abstract
Many complex generative systems use languages to create structured objects. We consider a model of random languages, defined by weighted context-free grammars. As the distribution of grammar weights broadens, a transition is found from a random phase, in which sentences are indistinguishable from noise, to an organized phase in which nontrivial information is carried. This marks the emergence of deep structure in the language, and can be understood by a competition between energy and entropy.
- Received 11 September 2018
- Revised 5 February 2019
DOI:https://doi.org/10.1103/PhysRevLett.122.128301
© 2019 American Physical Society
Physics Subject Headings (PhySH)
Focus
Learning Language Requires a Phase Transition
Published 29 March 2019
The transition a young child makes in acquiring complex language might be a kind of phase transition.
See more in Physics