Provably Efficient Model-free RL in Leader-Follower MDP with Linear Function Approximation

Keyphrases

• Efficient Model 100%
• Leader-follower 100%
• Regret Bounds 100%
• Model-free RL 100%
• Non-myopic 100%
• Linear Function Approximation 100%
• Bandit Feedback 100%
• Mechanism Design 50%
• Value Function 50%
• Performance Guarantee 50%
• Adaptation 50%
• Smart Grid 50%
• Security Design 50%
• Policy Making 50%
• Feedback Information 50%
• Multi-agent 50%
• CA Model 50%
• Interaction Type 50%
• Function Approximation 50%
• Number of States 50%
• Best Response 50%
• Number of Steps 50%
• Uniform Concentration 50%
• State Evolution 50%
• Joint Action 50%
• How to Learn 50%
• Softmax 50%
• Concentration Bounds 50%
• Sub-linear Regret 50%
• Feature Mapping 50%
• UCB Algorithm 50%
• Markov Games 50%
• Greedy Policy 50%
• Continuous State Space 50%
• RL Algorithm 50%
• Episodic MDP 50%

Computer Science

• Function Approximation 100%
• Linear Function 100%
• Smart Grid 50%
• Mechanism Design 50%
• Function Value 50%
• Performance Guarantee 50%
• multi agent 50%
• State Space 50%
• Feedback Information 50%
• Feature Mapping 50%
• Continuous Model 50%

Mathematics

• Function Value 100%
• Continuous Model 100%
• Total Number 100%
• Linear Function 100%
• Approximation Function 100%
• Joint Action 100%