Meta’s SPICE framework lets AI systems teach themselves to reason
via ai.meta.com
Short excerpt below. Read at the original source.
Researchers at Meta FAIR and the National University of Singapore have developed a new reinforcement learning framework for self-improving AI systems. Called Self-Play In Corpus Environments (SPICE), the framework pits two AI agents against each other, creating its own challenges and gradually improving without human supervision. While currently a proof-of-concept, this self-play mechanism could provide […]