Document worth reading: “Introduction to Multi-Armed Bandits”

Multi-armed bandits a simple nonetheless very extremely efficient framework for algorithms that make choices over time under uncertainty. An large physique of labor has gathered by way of the years, lined in numerous books and surveys. This e-book gives a additional introductory, textbook-like remedy of the subject. Each chapter tackles a selected line of labor, providing a self-contained, teachable technical introduction and a overview of the additional superior outcomes. The chapters are as follows: Stochastic bandits; Lower bounds; Bayesian Bandits and Thompson Sampling; Lipschitz Bandits; Full Feedback and Adversarial Costs; Adversarial Bandits; Linear Costs and Semi-bandits; Contextual Bandits; Bandits and Zero-Sum Games; Bandits with Knapsacks; Incentivized Exploration and Connections to Mechanism Design. Status of the manuscript: mainly full (modulo some sharpening), except for remaining chapter, which the creator plans to add over the next few months. Introduction to Multi-Armed Bandits