LATIDIA · Ciberseguridad

BARE-AI: Resiliencia a ataques de bit-flip en hardware de IA a través de monitores de rendimiento incorporados

arXiv: 2610.08739v1Tipo de anuncio: nuevo Resumen: las redes neuronales profundas (DNN) son parte integral de muchos sistemas críticos para la seguridad, pero siguen siendo altamente vulnerables a los ataques de volteo de bits (BFA), donde algunas perturbaciones a nivel de memoria

La noticia

arXiv:2610.08739v1 Announce Type: new Abstract: Deep Neural Networks (DNNs) are integral to many safety critical systems, yet they remain highly vulnerable to bit-flip attacks (BFAs), where a few memory level perturbations can drastically degrade accuracy. Existing defenses incur significant hardware overhead, depend on retraining, or fail against targeted flips. We propose BARE-AI, a runtime framework that detects, localizes, and mitigates BFAs during inference. BARE-AI introduces AI Performance Counters (APCs), lightweight hardware monitors in the accelerator datapath that capture per-layer activation statistics such as sparsity, entropy, kurtosis, and spectral shift. These are analyzed by the Predictive Unit for Layer Security Evaluation (PULSE), a compact detector trained offline as an ensemble of