When Samples Are Strategically Selected

Hanrui Zhang, Yu Cheng, Vincent Conitzer
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:7345-7353, 2019.

Abstract

In standard classification problems, the assumption is that the entity making the decision (the principal) has access to all the samples. However, in many contexts, she either does not have direct access to the samples, or can inspect only a limited set of samples and does not know which are the most relevant ones. In such cases, she must rely on another party (the agent) to either provide the samples or point out the most relevant ones. If the agent has a different objective, then the principal cannot trust the submitted samples to be representative. She must set a policy for how she makes decisions, keeping in mind the agent’s incentives. In this paper, we introduce a theoretical framework for this problem and provide key structural and computational results.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-zhang19c, title = {When Samples Are Strategically Selected}, author = {Zhang, Hanrui and Cheng, Yu and Conitzer, Vincent}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {7345--7353}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/zhang19c/zhang19c.pdf}, url = {https://proceedings.mlr.press/v97/zhang19c.html}, abstract = {In standard classification problems, the assumption is that the entity making the decision (the principal) has access to all the samples. However, in many contexts, she either does not have direct access to the samples, or can inspect only a limited set of samples and does not know which are the most relevant ones. In such cases, she must rely on another party (the agent) to either provide the samples or point out the most relevant ones. If the agent has a different objective, then the principal cannot trust the submitted samples to be representative. She must set a policy for how she makes decisions, keeping in mind the agent’s incentives. In this paper, we introduce a theoretical framework for this problem and provide key structural and computational results.} }
Endnote
%0 Conference Paper %T When Samples Are Strategically Selected %A Hanrui Zhang %A Yu Cheng %A Vincent Conitzer %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-zhang19c %I PMLR %P 7345--7353 %U https://proceedings.mlr.press/v97/zhang19c.html %V 97 %X In standard classification problems, the assumption is that the entity making the decision (the principal) has access to all the samples. However, in many contexts, she either does not have direct access to the samples, or can inspect only a limited set of samples and does not know which are the most relevant ones. In such cases, she must rely on another party (the agent) to either provide the samples or point out the most relevant ones. If the agent has a different objective, then the principal cannot trust the submitted samples to be representative. She must set a policy for how she makes decisions, keeping in mind the agent’s incentives. In this paper, we introduce a theoretical framework for this problem and provide key structural and computational results.
APA
Zhang, H., Cheng, Y. & Conitzer, V.. (2019). When Samples Are Strategically Selected. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:7345-7353 Available from https://proceedings.mlr.press/v97/zhang19c.html.

Related Material