Reversing Large Deep Learning AI Models - Yashodhan Vivek Mandke
Nullcon Goa 2025 · Main Stage
In this insightful Nullcon talk, Yashodhan Vivek Mandke delves into the critical, yet often overlooked, domain of **reversing deep learning (DL) models**. While the current discourse in AI security frequently centers on Large Language Models (LLMs) and peripheral attacks like prompt injection, Mandke argues that true security and robust defense necessitate a deeper understanding of the model's internal architecture and mathematical foundations. His presentation illuminates methods for dissecting complex AI models, exposing their core components, and identifying vulnerabilities that lie beyond superficial interactions.
AI review
Mandke brings a genuinely interesting framing — treating DL models as control systems and applying Bode/Root Locus analysis to identify weight vulnerability frequencies — that's fresher than the usual adversarial ML survey. The execution is thin, though: a PhD student demoing Netron on GoogleNet and hand-waving at sparsity metrics doesn't fully deliver on a premise that deserved much more rigor.