On Query-efficient Planning in MDPs under Linear Realizability of the Optimal State-value Function

Keyphrases

• Markov Decision Process 100%
• Efficient Planning 100%
• State-value Function 100%
• Value Function 75%
• Generative Models 50%
• Planners 50%
• Infinite Horizon 50%
• Optimal Policy 50%
• Action Sets 50%
• Deterministic Strategy 50%
• Feature Map 25%
• Constant Size 25%
• Local Access 25%
• Computation Cost 25%
• Feature Mapping 25%
• Optimal Value Function 25%
• Exponential number 25%
• Local Planning 25%
• Competing Values 25%
• Query Complexity 25%
• Random Transition 25%
• Fixed Horizon 25%
• Bounded Parameters 25%
• Complexity Guarantees 25%
• Action Value 25%
• Discounted Markov Decision Processes 25%

Computer Science

• Function Value 100%
• Markov Decision Process 100%
• Generative Model 33%
• Action-Set 33%
• And-States 16%
• Computation Cost 16%
• Feature Mapping 16%
• Exponential Number 16%
• Feature Map 16%

Mathematics

• Function Value 100%
• Markov Decision Process 100%
• Upper Bound 33%
• Optimal Policy 33%
• Polynomial 16%
• Open Question 16%
• Query Complexity 16%