causal

Human Prior Needed To Kickstart Ground-Truth RL

Imitation learning can leverage models to learn better from ground truth: models have been RLed to win IMO Gold and code entire applications from scratch—ground-truth examinations—but you couldn't RL a model to do these from scratch; you needed a reasonable prior over human data to kickstart the RL process.

causalpending

Speaker

Dwarkesh Patel

Evidence Quote

“You needed a reasonable prior over human data in order to kick start the RL process.”

Source

Some thoughts on the Sutton interview— Dwarkesh Patel Podcast
Created: 6/13/2026, 3:28:01 AM

Your notes

Loading notes...