Towards Q-learning the whittle index for restless bandits

Mathematics

• Q-learning 100%
• Recursion 34%
• Policy 23%
• Learning Systems 18%
• Reinforcement Learning 17%
• Multi-armed Bandit 16%
• Bandit Problems 16%
• Average Cost 15%
• Performance 14%
• Learning Algorithm 14%
• Infinite Horizon 13%
• Transition Matrix 13%
• Update 11%
• Controller 11%
• Markov Process 11%
• Optimal Solution 10%
• Optimal Control Problem 9%
• Knowledge 9%
• Subset 6%
• Approximation 6%

Engineering & Materials Science

• Markov processes 21%
• Learning systems 20%
• Learning algorithms 16%
• Controllers 10%
• Costs 7%