Virtual participation: Zoom details available here
Speaker: Naoise Holohan, IBM Research
Title: "Securing Floating-Point Arithmetic for Noise Addition"
Abstract: Floating-point arithmetic is ubiquitous across computing, with its wide range of values, large and small, making it the preferred tool for storing, analysing, and manipulating numerical data. Its flexibility comes at the cost of additional risks in some security/ privacy-aware settings. In this paper, we discuss the threat of infor- mation leakage caused by floating-point arithmetic when adding noise to sensitive values, which can allow the sensitive information to be recovered (e.g., in differential privacy). We present a solu- tion, Mantissa Bit Manipulation (MBM), that is orders of magnitude faster than the current state-of-the-art, applicable to most con- tinuous probability distributions and to all floating-point number formats.