Relational Boosted Bandits
This paper addresses the problem of applying contextual bandits to relational data for domains like social networks, offering an incremental solution for researchers and practitioners working with such data.
The authors propose Relational Boosted Bandits (RB2), a contextual bandit algorithm designed for relational domains like social networks, addressing the limitation of traditional contextual bandits that rely on attribute-value representations. RB2 is based on relational boosted trees and is demonstrated to be effective and interpretable on tasks such as link prediction, relational classification, and recommendations.
Contextual bandits algorithms have become essential in real-world user interaction problems in recent years. However, these algorithms rely on context as attribute value representation, which makes them unfeasible for real-world domains like social networks are inherently relational. We propose Relational Boosted Bandits(RB2), acontextual bandits algorithm for relational domains based on (relational) boosted trees. RB2 enables us to learn interpretable and explainable models due to the more descriptive nature of the relational representation. We empirically demonstrate the effectiveness and interpretability of RB2 on tasks such as link prediction, relational classification, and recommendations.