Join us on Wednesday 25 January at 10:30 in JFK, room E004/005, for a talk by Professor Lorenzo Cavallaro from University College London (UCL) as he presents Transcending Transcend: Revisiting Malware Classification in the Presence of Concept Drift.
Lorenzo Cavallaro is a Full Professor of Computer Science in the Department of Computer Science at UCL, where he leads the Systems Security Research Lab. His research focuses on understanding and improving the effectiveness of machine learning methods for systems security in the presence of adversaries. In particular, Lorenzo and his lab investigate the intertwined relationships of program analyses and machine learning and the implications they have towards realizing Trustworthy ML for Systems Security. Lorenzo publishes at and sits in the PC of top-tier conferences in computer security and received the Outstanding Paper Award at USENIX Security 2022.
In this talk, Professor Cavallaro will introduce TRANSCENDENT, a rejection framework built on Transcend, a classification with a rejection strategy based on conformal prediction theory. He will discuss the formal treatment of Transcend and showcase newly proposed conformal evaluators, with their different guarantees and computational properties. Attendees will learn how TRANSCENDENT outperforms state-of-the-art approaches while generalizing across various malware domains and classifiers. The insights presented in this talk will support both old and new empirical findings, making Transcend a sound and practical solution for the first time. The TRANSCENDENT framework will be released as open-source to aid the adoption of rejection strategies by the security community.
Don’t miss this opportunity to hear from an expert in the field and learn about the latest developments in malware classification.