FACADE: A Framework for Adversarial Circuit Anomaly Detection and Evaluation

Dhruv Bhandarkar Pai, Andres Carranza, Rylan Schaeffer, Arnuv Tandon, Sanmi Koyejo

Keywords: Interpretability, Anomaly Detection, Deep Learning, Probabilistic Models, Adversarial Robustness

TL;DR: We present FACADE, a novel probabilistic and geometric framework designed for unsupervised mechanistic anomaly detection in deep neural networks, aimed at advancing the understanding and mitigation of adversarial attacks.

Abstract: We present FACADE, a novel probabilistic and geometric framework designed for unsupervised mechanistic anomaly detection in deep neural networks. Its primary goal is advancing the understanding and mitigation of adversarial attacks. FACADE aims to generate probabilistic distributions over circuits, which provide critical insights to their contribution to changes in the manifold properties of pseudo-classes, or high-dimensional modes in activation space, yielding a powerful tool for uncovering and combating adversarial attacks. Our approach seeks to improve model robustness, enhance scalable model oversight, and demonstrates promising applications in real-world deployment settings.

Submission Number: 81