This work thus provides strong empirical evidence towards developing scaling laws for reinforcement learning.
