Optimal learning with non-Gaussian rewards

DOI10.1017/apr.2015.9zbMath1345.60039OpenAlexW2297707299MaRDI QIDQ2806349

Ilya O. Ryzhov, Zi Ding

Publication date: 17 May 2016

Published in: Advances in Applied Probability (Search for Journal in Brave)

Full work available at URL: https://projecteuclid.org/euclid.aap/1457466158

zbMATH Keywords

optimal stoppingLévy processmulti-armed banditpartial integro-differential equationoptimal learningGittins indicesnon-Gaussian rewardsprobabilistic interpolation

Mathematics Subject Classification ID

Related Items (3)

ON THE IDENTIFICATION AND MITIGATION OF WEAKNESSES IN THE KNOWLEDGE GRADIENT POLICY FOR MULTI-ARMED BANDITS ⋮ Undiscounted bandit games ⋮ Optimal stopping problems in Lévy models with random observations

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Conditional Lévy processes
• Optimal learning for sequential sampling with non-parametric beliefs
• Sur l'approximation des réduites. (On the approximation of residues)
• Convergence properties of the expected improvement algorithm with fixed mean and covariance functions
• Continuous multi-armed bandits and multiparameter processes
• The learning component of dynamic allocation indices
• On optimal stopping and free boundary problems
• Processes that can be embedded in Brownian motion
• Discrete multiarmed bandits and multiparameter processes
• Dynamic allocation problems in continuous time
• Optimal learning and experimentation in bandit problems.
• Lévy bandits: Multi-armed bandits driven by Lévy processes
• Convergence of values in optimal stopping and convergence of optimal stopping times
• Introductory lectures on fluctuations of Lévy processes with applications.
• The Knowledge Gradient Algorithm for a General Class of Online Learning Problems
• Consistency of Sequential Bayesian Sampling Policies
• Probability and Stochastics
• Multi‐Armed Bandit Allocation Indices
• Dynamic Pricing with a Prior on Market Response
• PROPERTIES OF THE GITTINS INDEX WITH APPLICATION TO OPTIMAL SCHEDULING
• Optimal investment and consumption with stochastic dividends
• Dynamic Assortment with Demand Learning for Seasonal Consumer Goods
• Stalking Information: Bayesian Inventory Management with Unobserved Lost Sales
• A Generalized Gittins Index for a Class of Multiarmed Bandits with General Resource Requirements
• A Knowledge-Gradient Policy for Sequential Information Collection
• The Multi-Armed Bandit Problem: Decomposition and Computation
• How Does the Value Function of a Markov Decision Process Depend on the Transition Probabilities?
• Sequential Testing Problems for Lévy Processes
• Bandit Problems with Lévy Processes
• Explicit Gittins Indices for a Class of Superdiffusive Processes
• Finite-time analysis of the multiarmed bandit problem

This page was built for publication: Optimal learning with non-Gaussian rewards