In the mathematical learning literature, reward-penalty rules have been studied in various decision-theoretic and game-theoretic contexts, the multi-armed bandit problem included. Here we propose an ...

How does a gambler maximize winnings from a row of slot machines? This is the inspiration for the "multi-armed bandit problem," a common task in reinforcement learning in which "agents" make choices ...

Thompson Sampling is an algorithm that can be used to analyze multi-armed bandit problems. Imagine you're in a casino standing in front of three slot machines. You have 10 free plays. Each machine ...

Imagine you’re a gambler and you’re standing in front of several slot machines. Your goal is to maximize your winnings, but you don’t actually know anything about the potential rewards offered by each ...