In-Place Zero-Space Memory Protection for CNN
This addresses the need for reliable CNN inference in autonomous vehicles and aerospace, offering a novel solution with no memory overhead, though it appears incremental as it builds on ECC concepts.
The paper tackled the problem of memory fault protection in CNNs for safety-critical applications by introducing an in-place zero-space ECC method with weight distribution-oriented training, achieving zero space cost without compromising reliability compared to traditional ECC.
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.