Sample Complexity of Reinforcement Learning using Linearly Combined Model Ensembles

Keyphrases

• Combined Model 100%
• Reinforcement Learning 100%
• Ensemble Model 100%
• Sample Complexity 100%
• Linear Combination 50%
• Challenging Problems 50%
• State Space 50%
• Real Environment 50%
• State Dependence 50%
• Model Selection 50%
• Reinforcement Learning Algorithm 50%
• Intelligent Behavior 50%
• Action Space 50%
• Exponential Lower Bounds 50%
• Near-optimal Policy 50%

Mathematics

• Polynomial 100%
• Approximates 100%
• Linear Combination 100%
• Model Selection 100%
• Optimal Policy 100%
• Action Space 100%
• Ensemble Model 100%

Computer Science

• Unknown Parameter 100%
• Reinforcement Learning 100%
• Efficient Algorithm 50%
• Linear Combination 50%
• Polynomial Complexity 50%
• Intelligent Behavior 50%