A Self-Play Posterior Sampling Algorithm for Zero-Sum Markov Games

Keyphrases

• Sampling Algorithm 100%
• Posterior Sampling 100%
• Markov Games 100%
• Self-play 100%
• Zero-sum Markov Game 100%
• Regret Bounds 33%
• Learning Settings 16%
• Complexity Measures 16%
• Uncertainty Principle 16%
• Multi-agent 16%
• Function Approximation 16%
• Frequentist 16%
• Reinforcement Learning 16%
• General Functioning 16%
• Regret 16%
• Bandit Learning 16%
• Optimism in the Face of Uncertainty 16%
• Decoupling Coefficient 16%

Mathematics

• Approximation Function 100%
• Frequentist 100%