Relational Boosted Bandits

Ashutosh Kakadiya, Sriraam Natarajan, Balaraman Ravindran

AAAI 2021 pp. 12123-12130

doi:10.1609/AAAI.V35I13.17439 /aaai/2021/kakadiya2021aaai-relational/

Abstract

Contextual bandits algorithms have become essential in real-world user interaction problems in recent years. However, these algorithms represent context as attribute value representation, which makes them infeasible for real world domains like social networks, which are inherently relational. We propose Relational Boosted Bandits (RB2), a contextual 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 recommendation.

PDF AAAI Semantic Scholar

Cite

Text

Kakadiya et al. "Relational Boosted Bandits." AAAI Conference on Artificial Intelligence, 2021. doi:10.1609/AAAI.V35I13.17439

Markdown

[Kakadiya et al. "Relational Boosted Bandits." AAAI Conference on Artificial Intelligence, 2021.](https://mlanthology.org/aaai/2021/kakadiya2021aaai-relational/) doi:10.1609/AAAI.V35I13.17439

BibTeX

@inproceedings{kakadiya2021aaai-relational,
  title     = {{Relational Boosted Bandits}},
  author    = {Kakadiya, Ashutosh and Natarajan, Sriraam and Ravindran, Balaraman},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2021},
  pages     = {12123-12130},
  doi       = {10.1609/AAAI.V35I13.17439},
  url       = {https://mlanthology.org/aaai/2021/kakadiya2021aaai-relational/}
}