TY - JOUR AU - Holohan, Naoise AU - Antonatos, Spiros AU - Braghin, Stefano AU - Mac Aonghusa, Pól PY - 2019/12/23 Y2 - 2024/03/28 TI - The Bounded Laplace Mechanism in Differential Privacy JF - Journal of Privacy and Confidentiality JA - JPC VL - 10 IS - 1 SE - TPDP 2018 DO - 10.29012/jpc.715 UR - https://journalprivacyconfidentiality.org/index.php/jpc/article/view/715 SP - AB - <p>The Laplace mechanism is the workhorse of differential privacy, applied to&nbsp;many instances where numerical data is processed. However, the Laplace mechanism can&nbsp;return semantically impossible values, such as negative counts, due to its infinite support.&nbsp;There are two popular solutions to this: (i) bounding/capping the output values and (ii)&nbsp;bounding the mechanism support. In this paper, we show that bounding the mechanism&nbsp;support, while using the parameters of the standard Laplace mechanism, does not typically preserve differential privacy. We also present a robust method to compute the optimal&nbsp;mechanism parameters to achieve differential privacy in such a setting.</p> ER -