A novel reinforcement learning algorithm is introduced for multiarmed restless bandits with
average reward, using the paradigms of Q-learning and Whittle index. Specifically, we …

Highlights•The study determines the optimal repair or replacement decision based on the
product lifecycle data.•A Markov Decision Process (MDP) framework is proposed for …

To overcome the curses of dimensionality and modeling of dynamic programming methods
to solve Markov decision process problems, reinforcement learning (RL) methods are …

We revisit the Whittle index policy for scheduling web crawlers for ephemeral content
proposed in Avrachenkov and Borkar, IEEE Trans. Control of Network Systems 5 (1), 2016 …

What poses a challenge for robot learning is the fact that considerable training samples and
time are required to find an optimal policy in the high-dimensional robot state space. To …

Hypertension is a crucial controllable risk factor of atherosclerotic cardiovascular disease, a
leading cause of death in the United States. While traditional analytic techniques may …