About this document

MAB-A: Engagement in Bandit Algorithms by ashwindandotiya is a document available to read on EtoBox.

The paper introduces a new model for multi-armed bandits (MAB) called MAB-A, which accounts for user abandonment in recommendation systems. It proposes two algorithms, ULCB and KL-ULCB, that optimize exploration and exploitation based on user engagement and past experiences, achieving logarithmic regret bounds. Simulation results demonstrate that these algorithms significantly outperform traditional MAB approaches in terms of regret reduction.

Author

ashwindandotiya

Language

EN