Incentivized Exploration for Multi-Armed Bandits Under Reward Drift
Abstract
We study incentivized exploration for the multi-armed bandit (MAB) problem where the players receive compensation for exploring arms other than the greedy choice and may provide biased feedback on reward. We seek to understand the impact of this drifted reward feedback by analyzing the performance of three instantiations of the incentivized MAB algorithm: UCB, ε-Greedy, and Thompson Sampling. Our results show that they all achieve O(log T) regret and compensation under the drifted reward, and are therefore effective in incentivizing exploration. Numerical examples are provided to complement the theoretical analysis.
Cite
Text
Liu et al. "Incentivized Exploration for Multi-Armed Bandits Under Reward Drift." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I04.5937Markdown
[Liu et al. "Incentivized Exploration for Multi-Armed Bandits Under Reward Drift." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/liu2020aaai-incentivized/) doi:10.1609/AAAI.V34I04.5937BibTeX
@inproceedings{liu2020aaai-incentivized,
title = {{Incentivized Exploration for Multi-Armed Bandits Under Reward Drift}},
author = {Liu, Zhiyuan and Wang, Huazheng and Shen, Fan and Liu, Kai and Chen, Lijun},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2020},
pages = {4981-4988},
doi = {10.1609/AAAI.V34I04.5937},
url = {https://mlanthology.org/aaai/2020/liu2020aaai-incentivized/}
}