Random and Adversarial Bit Error Robustness: Energy-Efficient and Secure DNN Accelerators. (arXiv:2104.08323v2 [cs.LG] UPDATED)

Deep neural network (DNN) accelerators received considerable attention in
recent years due to the potential to save energy compared to mainstream
hardware. Low-voltage operation of DNN accelerators allows to further reduce
energy consumption, however, causes bit-level failures in the memory storing
the quantized weights. Furthermore, DNN accelerators are vulnerable to
adversarial attacks on voltage controllers or individual bits. In this paper,
we show that a combination of robust fixed-point quantization, weight clipping,
as well as random bit error training (RandBET) …

arxiv dnn dnn accelerators energy error lg random robustness