POLY-HOOT: Monte-Carlo planning in continuous space MDPs with non-asymptotic analysis

Keyphrases

• Continuous Space 100%
• Non-asymptotic Analysis 100%
• Monte Carlo Tree Search 100%
• Monte-Carlo Planning 100%
• Optimistic Optimization 100%
• Finite Space 66%
• Convergence Guarantee 33%
• Optimization Algorithm 33%
• Robotics 33%
• Continuous States 33%
• State Action 33%
• Optimal Value Function 33%
• Bandits 33%
• Regret 33%
• Action Space 33%
• Bandit Problems 33%
• Nonasymptotic Convergence 33%
• Upper Confidence Bound 33%
• Polynomial Rate 33%
• AlphaGo Zero 33%
• Non-stationary Bandits 33%

Mathematics

• Monte Carlo 100%
• Asymptotic Analysis 100%
• Polynomial 50%
• Finite Space 33%
• Asymptotics 16%
• Function Value 16%
• Building Block 16%
• Action Space 16%
• Upper Confidence Bound 16%

Computer Science

• Continuous Space 100%
• Tree Search 100%
• Experimental Result 33%
• Optimization Algorithm 33%
• Function Value 33%
• Building-Blocks 33%
• Confidence Bound 33%