Behavior learning based on a policy gradient method: Separation of environmental dynamics and state values in policies

Mathematics

• Gradient Method 92%
• Policy 80%
• Learning 75%
• Value Function 39%
• Objective function 34%
• Rule Learning 31%
• Reinforcement Learning 31%
• Reuse 28%
• Pursuit 25%
• Simulation Experiment 23%
• Decision problem 21%
• Gradient 15%

Engineering & Materials Science

• Gradient methods 100%
• Reinforcement learning 20%
• Experiments 7%