In-Place Zero-Space Memory Protection for CNN
Hui Guan · Lin Ning · Zhen Lin · Xipeng Shen · Huiyang Zhou · Seung-Hwan Lim

Tue Dec 10th 05:30 -- 07:30 PM @ East Exhibition Hall B + C #90

Convolutional Neural Networks (CNN) are being actively explored for safety-critical applications such as autonomous vehicles and aerospace, where it is essential to ensure the reliability of inference results in the presence of possible memory faults. Traditional methods such as error correction codes (ECC) and Triple Modular Redundancy (TMR) are CNN-oblivious and incur substantial memory overhead and energy cost. This paper introduces in-place zero-space ECC assisted with a new training scheme weight distribution-oriented training. The new method provides the first known zero space cost memory protection for CNNs without compromising the reliability offered by traditional ECC.

Author Information

Hui Guan (North Carolina State University)
Lin Ning (NCSU)
Zhen Lin (NCSU)
Xipeng Shen (North Carolina State University)
Huiyang Zhou (NCSU)
Seung-Hwan Lim (Oak Ridge National Laboratory)