Tag: #PPO

← All posts

• Policy Gradients and PPO: Why We Clip Ratios, Advantage Functions and GAE, Value Functions, Entropy

  Policy gradients increase the probability of actions that performed better than usual, and PPO reuses the same data multiple times by clipping the probability ratio between the new and old policies so they do not move to

  2026-10-18 · 12 min read #탁구 로봇#강화학습#PPO#정책 경사#GAE

• Table Tennis Arm Reinforcement Learning: Past the Baseline, but Most Gains Were Not from Learning

  We trained two types of reinforcement learning using the same ball and robot as five classical control baselines. Hybrid RL, which only selects striking parameters, significantly outperformed the baselines (61.6, 59.2, a

  2026-10-17 · 15 min read #탁구 로봇#강화학습#PPO#MuJoCo Warp#하이브리드 제어