Constrained differentiable cross-entropy method for safe model-based reinforcement learning

Keyphrases

• Reinforcement Learning Agent 100%
• Cross-entropy 100%
• Model Predictive Control 66%
• Reinforcement Learning Algorithm 66%
• Safety Concerns 33%
• System Dynamics 33%
• Optimization Problem 33%
• Extra Assumption 33%
• Optimal Policy 33%
• Trial-and-error 33%
• Constraint Violation 33%
• Control Framework 33%
• Reinforcement Learning 33%
• Mismatch Problem 33%

Computer Science

• Reinforcement Learning 100%
• Learning Agent 50%
• Predictive Model 33%
• Optimization Problem 16%
• System Dynamics 16%
• Energy Optimization 16%
• Constraint Violation 16%
• Control Framework 16%