Learning End-to-End Visual Servoing Using an Improved Soft Actor-Critic Approach With Centralized Novelty Measurement

Computer Science

• Reinforcement Learning 100%
• Neural Network 50%
• Maximum Entropy 25%
• Extracted Feature 25%
• Training Process 25%
• Measurement Method 25%
• Learning Agent 25%
• temporal difference 25%
• Completion Rate 25%