Feargus Pendlebury, Royal Holloway, University of London
From 12:30 until 13:30
At Zoom: https://ethz.zoom.us/j/64900469091
https://ethz.zoom.us/j/64900469091
Abstract:
With the growing processing power of computing systems and the increased availability of massive datasets, machine learning algorithms have led to major breakthroughs in many different areas. This development has influenced computer security, inspiring many learning-based security systems, such as for malware detection, vulnerability discovery, and binary code analysis. Despite great potential, machine learning in security is prone to subtle pitfalls that undermine its performance and render learning-based systems potentially unsuitable for security tasks and practical deployment.
In this seminar, we look at common pitfalls in the design, implementation, and evaluation of learning-based security systems which we have identified across 30 papers from top-tier security conferences within the past decade. We further examine how individual pitfalls can lead to unrealistic and misleading results through a set of case studies and, as a remedy, derive actionable recommendations for avoiding them. This seminar is based on work set to appear at USENIX Security 2022.
Join the Zoom meeting at 12:30 on Thursday, October 21st: https://ethz.zoom.us/j/64900469091