No more hand-tuning rewards: Masked constrained policy optimization for safe reinforcement learning

Computer Science

• Reinforcement Learning 100%
• Optimization Policy 100%
• Learning Process 50%
• Optimization Problem 25%
• Learning Problem 25%
• Expert Knowledge 25%
• Terminal State 25%

Engineering

• Reinforcement Learning 100%
• Engineering 25%
• Search Method 25%
• Optimisation Problem 25%
• Maximization 25%
• Reward Model 25%
• Performance Trade 25%
• Safety Performance 25%

Chemical Engineering

• Reinforcement Learning 100%