TY - JOUR
AU - Whitehill, Jacob
PY - 2019/07/17
Y2 - 2022/09/25
TI - How Does Knowledge of the AUC Constrain the Set of Possible Ground-Truth Labelings?
JF - Proceedings of the AAAI Conference on Artificial Intelligence
JA - AAAI
VL - 33
IS - 01
SE - AAAI Technical Track: Machine Learning
DO - 10.1609/aaai.v33i01.33015425
UR - https://ojs.aaai.org/index.php/AAAI/article/view/4482
SP - 5425-5432
AB - <p>Recent work on privacy-preserving machine learning has considered how datamining competitions such as Kaggle could potentially be “hacked”, either intentionally or inadvertently, by using information from an oracle that reports a classifier’s accuracy on the test set (Blum and Hardt 2015; Hardt and Ullman 2014; Zheng 2015; Whitehill 2016). For binary classification tasks in particular, one of the most common accuracy metrics is the Area Under the ROC Curve (AUC), and in this paper we explore the mathematical structure of how the AUC is computed from an <em>n</em>-vector of real-valued “guesses” with respect to the ground-truth labels. Under the assumption of perfect knowledge of the test set AUC <em>c</em>=<em>p/q</em>, we show how knowing <em>c</em> constrains the set W of possible ground-truth labelings, and we derive an algorithm both to compute the exact number of such labelings and to enumerate efficiently over them. We also provide empirical evidence that, surprisingly, the number of compatible labelings can actually <em>decrease</em> as <em>n</em> grows, until a test set-dependent threshold is reached. Finally, we show how W can be efficiently whittled down, through pairs of oracle queries, to infer all the groundtruth test labels with complete certainty.</p>
ER -