Not All Transitions Matter: Evidence from PPO

In three linesarXiv paper showing consecutive transitions in on-policy RL are redundant and cause hidden instability. Randomly dropping 25% of transitions in PPO stabilizes training (KL divergence, entropy, value estimates) without degrading rewards, across CartPole, Acrobot, LunarLander, HalfCheetah, Hopper.

Summary generated by Claude — human-verified