Timezone: »

 
Spotlight
Strategic Classification in the Dark
Ganesh Ghalme · Vineet Nair · Itay Eilat · Inbal Talgam-Cohen · Nir Rosenfeld

Thu Jul 22 05:25 AM -- 05:30 AM (PDT) @ None
in Fairness »

Strategic classification studies the interaction between a classification rule and the strategic agents it governs. Agents respond by manipulating their features, under the assumption that the classifier is known. However, in many real-life scenarios of high-stake classification (e.g., credit scoring), the classifier is not revealed to the agents, which leads agents to attempt to learn the classifier and game it too. In this paper we generalize the strategic classification model to such scenarios and analyze the effect of an unknown classifier. We define the ''price of opacity'' as the difference between the prediction error under the opaque and transparent policies, characterize it, and give a sufficient condition for it to be strictly positive, in which case transparency is the recommended policy. Our experiments show how Hardt et al.’s robust classifier is affected by keeping agents in the dark.

Keywords: Fairness, Accountability, and Transparency · Social Aspects of Machine Learning
[ Paper ]

Author Information

Ganesh Ghalme (Technion- Israel Institute of Technolgy, Haifa)
Vineet Nair (Technion)
Itay Eilat (Technion)
Inbal Talgam-Cohen (Technion)
Nir Rosenfeld (Harvard)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors