On the Convergence Rate of Off-Policy Policy Optimization Methods with Density-Ratio Correction

Keyphrases

• Convergence Rate 100%
• Optimization Methods 100%
• Policy Optimization 100%
• Ratio Correction 100%
• Convergence Guarantee 33%
• Convergence Property 33%
• Stationary Point 33%
• Learning Objectives 33%
• Optimization Algorithm 33%
• Function Approximation 33%
• State Action 33%
• Minimax Problem 33%
• Total Complexity 33%
• MaxMax 33%
• Finite-time Convergence 33%
• Actor-critic Algorithm 33%
• Actor-critic 33%
• On-policy 33%
• Policy Actors 33%
• Action Density 33%
• SREDA 33%

Mathematics

• Convergence Rate 100%
• Convergence Property 33%
• Min-Max 33%
• Approximation Function 33%
• Stationary Point 33%
• Objective Function 33%
• Finite Time 33%