A Few Expert Queries Suffices for Sample-Efficient RL with Resets and Linear Value Approximation

Keyphrases

• Value Function 100%
• Delphi Method 100%
• Value Function Approximation 100%
• Sample Complexity 100%
• Sample-Efficient Reinforcement Learning 100%
• Linear Value 100%
• Weak Assumption 50%
• Computationally Efficient Algorithms 50%
• Learning Approaches 50%
• Generative Models 50%
• Oracle 50%
• Expert Demonstration 50%
• Imitation Learning 50%
• Reinforcement Learning Approach 50%
• Optimal Value Function 50%
• Suboptimal Policies 50%
• Polynomially Bounded 50%
• Interactive Demonstration 50%
• Linear Factor 50%

Computer Science

• Approximation (Algorithm) 100%
• Function Value 100%
• Reinforcement Learning 100%
• Learning Approach 66%
• Efficient Algorithm 33%
• Generative Model 33%
• Minimal Amount 33%

Mathematics

• Function Value 100%
• Weaker Assumption 33%
• Worst Case 33%
• Exploratory 33%
• Linear Factor 33%