ERASER: Machine Unlearning in MLaaS via an Inference Serving-Aware Approach

Over the past few years, Machine Learning-as-a-Service (MLaaS) has received a surging demand for supporting Machine Learning-driven services to offer revolutionized user experience across diverse application areas. MLaaS provides inference service with low inference latency to application users based on an ML model trained using a dataset collected from numerous individual data owners. Recently, for the sake of data owners' privacy and to comply with the "right to be forgotten (RTBF)" as enacted by data protection legislation, many machine unlearning methods have been proposed to remove data owners' data from trained models upon their unlearning requests. However, despite their promising efficiency, almost all existing machine unlearning methods handle unlearning requests in a manner that is independent of inference requests, which unfortunately introduces new security and privacy vulnerabilities for machine unlearning in MLaaS. In this paper, we propose the ERASER framework for machinE unleaRning in MLaAS via an inferencE seRving-aware approach. ERASER proposes a novel certified inference consistency mechanism that reduces inference latency by selectively postponing unlearning execution incurred by unlearning requests from data owners, while strictly adhering to the RTBF principle. ERASER offers three groups of design choices to allow for tailor-made variants that best suit the specific environments and preferences of different MLaaS systems. Extensive empirical evaluations across various settings confirm ERASER's effectiveness, e.g., it can effectively save up to 99% of inference latency and 31% of computation overhead over the inference-oblivion baseline.