ALARM: Active LeArning of Rowhammer Mitigations
This addresses a security issue for DRAM customers by providing a method to assess mitigations, though it is incremental as it builds on existing active learning techniques.
The paper tackled the problem of evaluating the efficacy of undisclosed Rowhammer mitigations in DRAM by developing a tool that uses active learning to automatically infer mitigation parameters against synthetic models, achieving automated inference without disclosing specific numerical results.
Rowhammer is a serious security problem of contemporary dynamic random-access memory (DRAM) where reads or writes of bits can flip other bits. DRAM manufacturers add mitigations, but don't disclose details, making it difficult for customers to evaluate their efficacy. We present a tool, based on active learning, that automatically infers parameter of Rowhammer mitigations against synthetic models of modern DRAM.