Derivative-Free Policy Optimization for Linear Risk-Sensitive and Robust Control Design: Implicit Regularization and Sample Complexity

Keyphrases

• Policy Optimization 100%
• Robust Control Design 100%
• Risk Control 100%
• Derivative-free Method 100%
• Policy Gradient Method 100%
• Sample Complexity 100%
• Free Policy 100%
• Implicit Regularization 100%
• Linear Quadratic 66%
• Global Convergence 66%
• Nonconvex 33%
• Sampled-data Systems 33%
• Controller 33%
• Continuous Space 33%
• Disturbance Attenuation 33%
• System Trajectory 33%
• Dynamic Games 33%
• Robust Control 33%
• Robust Controller 33%
• Risk Sensitivity 33%
• Zero-sum 33%
• Quadratic Dynamics 33%
• Complexity Results 33%
• Reinforcement Learning 33%
• Convergence Theory 33%
• Risk-sensitive 33%
• Linear Exponential Quadratic Gaussian 33%
• Critical Control Systems 33%
• Nonconcave 33%
• Min-max Optimization 33%
• Learning Phase 33%
• Global Sample 33%
• Convergence Complexity 33%
• Multi-agent Reinforcement Learning 33%
• Continuous Control Tasks 33%
• Safety-critical Control 33%
• Regularization Property 33%
• Direct Policy Search 33%

Computer Science

• Regularization 100%
• Optimization Policy 100%
• Gradient Method 100%
• Global Convergence 66%
• Optimization Problem 33%
• Continuous Space 33%
• System Trajectory 33%
• Risk Sensitivity 33%
• Reinforcement Learning 33%
• Learning Phase 33%
• Complexity Result 33%
• Critical Control 33%
• Continuous Control 33%
• Multi-Agent Reinforcement Learning 33%

Mathematics

• Regularization 100%
• Robust Control 100%
• Gaussian Distribution 50%
• Minimax 50%
• Control System 50%
• Dynamic Game 50%
• Safety-Critical Control 50%