On the (In)feasibility of ML Backdoor Detection as an Hypothesis Testing Problem

arXiv:2402.16926v1 Announce Type: cross
Abstract: We introduce a formal statistical definition for the problem of backdoor detection in machine learning systems and use it to analyze the feasibility of such problems, providing evidence for the utility and applicability of our definition. The main contributions of this work are an impossibility result and an achievability result for backdoor detection. We show a no-free-lunch theorem, proving that universal (adversary-unaware) backdoor detection is impossible, except for very small alphabet sizes. Thus, we argue, …

abstract analyze arxiv backdoor cs.ai cs.cr cs.lg definition detection evidence hypothesis learning systems machine machine learning statistical stat.ml systems testing type utility work