Bandits

(à venir)

Lectures

• Modeles de bandit : une histoire bayésienne et fréquentiste

• Bandit theory, part I

• Bandit theory, part II

• Kernel-based methods for bandit convex optimization, part 1

• Kernel-based methods for bandit convex optimization, part 2

• Kernel-based methods for bandit convex optimization, part 3

• Learning to Interact (John Langford)

• Batch Learning from Logged Bandit Feedback through Counterfactual Risk Minimization

• Stochastic Structured Prediction under Bandit Feedback

• Thompson sampling with the online bootstrap (à lire)

• Trial without Error: Towards Safe Reinforcement Learning via Human Intervention

• Corrupt Bandits for Preserving Local Privacy

• Multi-Player Bandits Revisited

• Learning the distribution with largest mean: two bandit frameworks

• Analyse de stratégies bayésiennes et fréquentistes pour l’allocation séquentielle de ressources (thèse)

Modules

• SMPyBandits, SMPyBandits, a Research Framework for Single and Multi-Players Multi-Arms Bandits Algorithms in Python