1.4 RL

• 1.4.0 RL基础
  • Actor-Critic
  • Bellman方程
  • GAE
  • MDP
  • Model-free & Model-based
  • On-policy & Off-policy
  • Policy Gradient
  • Reward设计
  • Value & Q & Advantage
• 1.4.1 Online RL
  • A2C & A3C
  • DDPG
  • HIL-SERL
  • PPO
  • SAC
  • TD3
• 1.4.2 VLA Post-training
  • GRPO for VLA
  • PPO for VLA
  • RLinf-VLA