Learning to Incentivize Information Acquisition: Proper Scoring Rules Meet Principal-Agent Model

Siyu Chen, Jibang Wu, Yifan Wu, Zhuoran Yang
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:5194-5218, 2023.

Abstract

We study the incentivized information acquisition problem, where a principal hires an agent to gather information on her behalf. Such a problem is modeled as a Stackelberg game between the principal and the agent, where the principal announces a scoring rule that specifies the payment, and then the agent then chooses an effort level that maximizes her own profit and reports the information. We study the online setting of such a problem from the principal’s perspective, i.e., designing the optimal scoring rule by repeatedly interacting with the strategic agent. We design a provably sample efficient algorithm that tailors the UCB algorithm (Auer et al., 2002) to our model, which achieves a $\mathcal{O} (K^2\cdot T^{2/3})$ regret after $T$ iterations, where $K$ is the number of effort levels of the agent. Our algorithm features a delicate estimation procedure for the optimal profit of the principal, and a conservative correction scheme that ensures the desired agent’s actions are incentivized. Furthermore, a key feature of our regret bound is that it is independent of the number of states of the environment.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-chen23ah, title = {Learning to Incentivize Information Acquisition: Proper Scoring Rules Meet Principal-Agent Model}, author = {Chen, Siyu and Wu, Jibang and Wu, Yifan and Yang, Zhuoran}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {5194--5218}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/chen23ah/chen23ah.pdf}, url = {https://proceedings.mlr.press/v202/chen23ah.html}, abstract = {We study the incentivized information acquisition problem, where a principal hires an agent to gather information on her behalf. Such a problem is modeled as a Stackelberg game between the principal and the agent, where the principal announces a scoring rule that specifies the payment, and then the agent then chooses an effort level that maximizes her own profit and reports the information. We study the online setting of such a problem from the principal’s perspective, i.e., designing the optimal scoring rule by repeatedly interacting with the strategic agent. We design a provably sample efficient algorithm that tailors the UCB algorithm (Auer et al., 2002) to our model, which achieves a $\mathcal{O} (K^2\cdot T^{2/3})$ regret after $T$ iterations, where $K$ is the number of effort levels of the agent. Our algorithm features a delicate estimation procedure for the optimal profit of the principal, and a conservative correction scheme that ensures the desired agent’s actions are incentivized. Furthermore, a key feature of our regret bound is that it is independent of the number of states of the environment.} }
Endnote
%0 Conference Paper %T Learning to Incentivize Information Acquisition: Proper Scoring Rules Meet Principal-Agent Model %A Siyu Chen %A Jibang Wu %A Yifan Wu %A Zhuoran Yang %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-chen23ah %I PMLR %P 5194--5218 %U https://proceedings.mlr.press/v202/chen23ah.html %V 202 %X We study the incentivized information acquisition problem, where a principal hires an agent to gather information on her behalf. Such a problem is modeled as a Stackelberg game between the principal and the agent, where the principal announces a scoring rule that specifies the payment, and then the agent then chooses an effort level that maximizes her own profit and reports the information. We study the online setting of such a problem from the principal’s perspective, i.e., designing the optimal scoring rule by repeatedly interacting with the strategic agent. We design a provably sample efficient algorithm that tailors the UCB algorithm (Auer et al., 2002) to our model, which achieves a $\mathcal{O} (K^2\cdot T^{2/3})$ regret after $T$ iterations, where $K$ is the number of effort levels of the agent. Our algorithm features a delicate estimation procedure for the optimal profit of the principal, and a conservative correction scheme that ensures the desired agent’s actions are incentivized. Furthermore, a key feature of our regret bound is that it is independent of the number of states of the environment.
APA
Chen, S., Wu, J., Wu, Y. & Yang, Z.. (2023). Learning to Incentivize Information Acquisition: Proper Scoring Rules Meet Principal-Agent Model. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:5194-5218 Available from https://proceedings.mlr.press/v202/chen23ah.html.

Related Material