Strategic Representation

Vineet Nair, Ganesh Ghalme, Inbal Talgam-Cohen, Nir Rosenfeld
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:16331-16352, 2022.

Abstract

Humans have come to rely on machines for reducing excessive information to manageable representations. But this reliance can be abused – strategic machines might craft representations that manipulate their users. How can a user make good choices based on strategic representations? We formalize this as a learning problem, and pursue algorithms for decision-making that are robust to manipulation. In our main setting of interest, the system represents attributes of an item to the user, who then decides whether or not to consume. We model this interaction through the lens of strategic classification (Hardt et al. 2016), reversed: the user, who learns, plays first; and the system, which responds, plays second. The system must respond with representations that reveal ‘nothing but the truth’ but need not reveal the entire truth. Thus, the user faces the problem of learning set functions under strategic subset selection, which presents distinct algorithmic and statistical challenges. Our main result is a learning algorithm that minimizes error despite strategic representations, and our theoretical analysis sheds light on the trade-off between learning effort and susceptibility to manipulation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-nair22a, title = {Strategic Representation}, author = {Nair, Vineet and Ghalme, Ganesh and Talgam-Cohen, Inbal and Rosenfeld, Nir}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {16331--16352}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/nair22a/nair22a.pdf}, url = {https://proceedings.mlr.press/v162/nair22a.html}, abstract = {Humans have come to rely on machines for reducing excessive information to manageable representations. But this reliance can be abused – strategic machines might craft representations that manipulate their users. How can a user make good choices based on strategic representations? We formalize this as a learning problem, and pursue algorithms for decision-making that are robust to manipulation. In our main setting of interest, the system represents attributes of an item to the user, who then decides whether or not to consume. We model this interaction through the lens of strategic classification (Hardt et al. 2016), reversed: the user, who learns, plays first; and the system, which responds, plays second. The system must respond with representations that reveal ‘nothing but the truth’ but need not reveal the entire truth. Thus, the user faces the problem of learning set functions under strategic subset selection, which presents distinct algorithmic and statistical challenges. Our main result is a learning algorithm that minimizes error despite strategic representations, and our theoretical analysis sheds light on the trade-off between learning effort and susceptibility to manipulation.} }
Endnote
%0 Conference Paper %T Strategic Representation %A Vineet Nair %A Ganesh Ghalme %A Inbal Talgam-Cohen %A Nir Rosenfeld %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-nair22a %I PMLR %P 16331--16352 %U https://proceedings.mlr.press/v162/nair22a.html %V 162 %X Humans have come to rely on machines for reducing excessive information to manageable representations. But this reliance can be abused – strategic machines might craft representations that manipulate their users. How can a user make good choices based on strategic representations? We formalize this as a learning problem, and pursue algorithms for decision-making that are robust to manipulation. In our main setting of interest, the system represents attributes of an item to the user, who then decides whether or not to consume. We model this interaction through the lens of strategic classification (Hardt et al. 2016), reversed: the user, who learns, plays first; and the system, which responds, plays second. The system must respond with representations that reveal ‘nothing but the truth’ but need not reveal the entire truth. Thus, the user faces the problem of learning set functions under strategic subset selection, which presents distinct algorithmic and statistical challenges. Our main result is a learning algorithm that minimizes error despite strategic representations, and our theoretical analysis sheds light on the trade-off between learning effort and susceptibility to manipulation.
APA
Nair, V., Ghalme, G., Talgam-Cohen, I. & Rosenfeld, N.. (2022). Strategic Representation. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:16331-16352 Available from https://proceedings.mlr.press/v162/nair22a.html.

Related Material