Gaussian process temporal-difference learning with Scalability and worst-case performance guarantees

Keyphrases

• Adversarial Setting 50%
• Bellman 50%
• Benchmark Problems 50%
• Gaussian Process 100%
• Gaussian Process Prior 50%
• Long-term Rewards 50%
• Model Assumptions 50%
• Performance Guarantee 100%
• Policy Evaluation 100%
• Process Perspective 50%
• Random Features 50%
• Reinforcement Learning 50%
• Reward Prediction Error 50%
• Scalable Approaches 50%
• State Space 50%
• Temporal Difference Learning 100%
• Value Function 100%
• Value Function Approximation 50%
• Worst-case Analysis 50%
• Worst-case Performance 100%

Computer Science

• Adversarial Setting 50%
• Benchmark Problem 50%
• Function Approximation 50%
• Performance Guarantee 100%
• Prediction Error 50%
• Process Perspective 50%
• Reinforcement Learning 50%
• State Space 50%
• temporal difference learning 100%