Near-optimal polynomial time algorithm for learning in certain classes of stochastic games

Keyphrases

• Adversary 50%
• Exploration-exploitation 25%
• Game Matrix 50%
• Given Information 25%
• Markov Decision Process 25%
• Near-optimal 100%
• Optimal Behavior 25%
• Optimal Return 25%
• Polynomial Time 25%
• Polynomial-time Algorithm 100%
• Reinforcement Learning 25%
• Repeated Games 25%
• Singh 25%
• Single Controller 50%
• Stochastic Games 100%
• Withholding 25%

Computer Science

• Markov Decision Process 50%
• Polynomial Time 50%
• Polynomial Time Algorithm 100%
• Reinforcement Learning 50%