Dataset Characteristics and Their Impact on Offline Policy Learning of Contextual Multi-Armed Bandits

Piotr Januszewski, Dominik Grzegorzek, Paweł Czarnul

2024

Abstract

The Contextual Multi-Armed Bandits (CMAB) framework is pivotal for learning to make decisions. However, due to challenges in deploying online algorithms, there is a shift towards offline policy learning, which relies on pre-existing datasets. This study examines the relationship between the quality of these datasets and the performance of offline policy learning algorithms, specifically, Neural Greedy and NeuraLCB. Our results demonstrate that NeuraLCB can learn from various datasets, while Neural Greedy necessitates extensive coverage of the action-space for effective learning. Moreover, the way data is collected significantly affects offline methods’ efficiency. This underscores the critical role of dataset quality in offline policy learning.

Paper Citation

in Harvard Style

Januszewski P., Grzegorzek D. and Czarnul P. (2024). Dataset Characteristics and Their Impact on Offline Policy Learning of Contextual Multi-Armed Bandits. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 87-98. DOI: 10.5220/0012311000003636

in Bibtex Style

@conference{icaart24,
author={Piotr Januszewski and Dominik Grzegorzek and Paweł Czarnul},
title={Dataset Characteristics and Their Impact on Offline Policy Learning of Contextual Multi-Armed Bandits},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2024},
pages={87-98},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012311000003636},
isbn={978-989-758-680-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - Dataset Characteristics and Their Impact on Offline Policy Learning of Contextual Multi-Armed Bandits
SN - 978-989-758-680-4
AU - Januszewski P.
AU - Grzegorzek D.
AU - Czarnul P.
PY - 2024
SP - 87
EP - 98
DO - 10.5220/0012311000003636
PB - SciTePress