Pure Exploration in Asynchronous Federated Bandits

Zichen Wang, Chuanhao Li, Chenyu Song, Lianghui Wang, Quanquan Gu, Huazheng Wang
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, PMLR 244:3540-3570, 2024.

Abstract

We study the federated pure exploration problem of multi-armed bandits and linear bandits, where $M$ agents cooperatively identify the best arm via communicating with the central server. To enhance the robustness against latency and unavailability of agents that are common in practice, we propose the first federated asynchronous multi-armed bandit and linear bandit algorithms for pure exploration with fixed confidence. Our theoretical analysis shows the proposed algorithms achieve near-optimal sample complexities and efficient communication costs in a fully asynchronous environment. Moreover, experimental results based on synthetic and real-world data empirically elucidate the effectiveness and communication cost-efficiency of the proposed algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v244-wang24c, title = {Pure Exploration in Asynchronous Federated Bandits}, author = {Wang, Zichen and Li, Chuanhao and Song, Chenyu and Wang, Lianghui and Gu, Quanquan and Wang, Huazheng}, booktitle = {Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence}, pages = {3540--3570}, year = {2024}, editor = {Kiyavash, Negar and Mooij, Joris M.}, volume = {244}, series = {Proceedings of Machine Learning Research}, month = {15--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v244/main/assets/wang24c/wang24c.pdf}, url = {https://proceedings.mlr.press/v244/wang24c.html}, abstract = {We study the federated pure exploration problem of multi-armed bandits and linear bandits, where $M$ agents cooperatively identify the best arm via communicating with the central server. To enhance the robustness against latency and unavailability of agents that are common in practice, we propose the first federated asynchronous multi-armed bandit and linear bandit algorithms for pure exploration with fixed confidence. Our theoretical analysis shows the proposed algorithms achieve near-optimal sample complexities and efficient communication costs in a fully asynchronous environment. Moreover, experimental results based on synthetic and real-world data empirically elucidate the effectiveness and communication cost-efficiency of the proposed algorithms.} }
Endnote
%0 Conference Paper %T Pure Exploration in Asynchronous Federated Bandits %A Zichen Wang %A Chuanhao Li %A Chenyu Song %A Lianghui Wang %A Quanquan Gu %A Huazheng Wang %B Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Negar Kiyavash %E Joris M. Mooij %F pmlr-v244-wang24c %I PMLR %P 3540--3570 %U https://proceedings.mlr.press/v244/wang24c.html %V 244 %X We study the federated pure exploration problem of multi-armed bandits and linear bandits, where $M$ agents cooperatively identify the best arm via communicating with the central server. To enhance the robustness against latency and unavailability of agents that are common in practice, we propose the first federated asynchronous multi-armed bandit and linear bandit algorithms for pure exploration with fixed confidence. Our theoretical analysis shows the proposed algorithms achieve near-optimal sample complexities and efficient communication costs in a fully asynchronous environment. Moreover, experimental results based on synthetic and real-world data empirically elucidate the effectiveness and communication cost-efficiency of the proposed algorithms.
APA
Wang, Z., Li, C., Song, C., Wang, L., Gu, Q. & Wang, H.. (2024). Pure Exploration in Asynchronous Federated Bandits. Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 244:3540-3570 Available from https://proceedings.mlr.press/v244/wang24c.html.

Related Material