Nowadays, as Deep Neural Networks (DNNs) become ubiquitous in mission-critical domains (e.g., automatic driving systems), ensuring the reliable execution of domain-specific DNN accelerators in the presence of hardware faults is increasingly essential. However, recent fault tolerance attempts either suffer from expensive performance overhead or are limited in scalability by specific fault modes. This paper proposes HAp-FT, a Hybrid Approximate fault Tolerance framework that integrates both proactive error alleviation and efficient error detection design philosophies. First, considering the vertical accumulation execution paradigm of the systolic array, HAp-FT proactively transfers the reliability risk of vulnerable weight to neighboring Processing Elements (PEs) for advance fault effect alleviation. Then, leveraging the inter-filter similarity characteristics, HAp-FT remaps similar filters to adjacent columns of PEs to achieve real-time intragroup approximate error detection. Experimental results exhibit that HAp-FT can recover 98.38%95.74% accuracy degradation incurred by transient/permanent hardware faults while only introducing 0.18% performance overhead, 2.69% extra area, and 2.38% energy overhead. Moreover, to satisfy the reliability requirements of various application scenarios, HAp-FT is also portable to systolic arrays under all dataflow strategies and adaptable quantized models.

The application of INT quantization allows CNN model size and computation to be reduced exponentially, making extreme quantization a hot research topic. However, firstly the decrease in bitwidth increases the probability of error in critical bits, which is unacceptable for safety-critical scenarios that require high resilience of CNNs. Secondly, extreme quantization often requires retraining with high overhead. Considering the higher tolerance of errors in high-bitwidth INTs and the local resilience differences of CNNs, we use mixed precision quantization (MPQ) to trade off resilience and compression effects. Existing MPQ methods mostly focus on compression as the first goal and lack indicators of resilience or vulnerability. Based on a large number of vulnerability analyses, we propose to use the maximum bit average offset and the approximation parameter average quantization error to measure layer vulnerability. And we measure the vulnerability of the entire CNN in terms of the accumulation of layer offsets. Accordingly, we design a complete vulnerability-aware MPQ (VulnA-MPQ) framework. From the perspective of minimizing vulnerability and compression ratio, VulnA-MPQ is performed on multiple CNNs using genetic algorithm (GA) with low overhead. The CNNs only need to be fine-tuned to restore accuracy after VulnA-MPQ. Experiments demonstrate that VulnA-MPQ not only maintains accuracy and compression ratio better but also improves CNNs fault tolerance. Within the 2% accuracy threshold, the random error data tolerated by VGG16 weights and activations are improved by 33%. The number of random errors tolerated by MobileNetV3-large weights is improved by 8.87 times, and that of activations is improved by 4.64 times.

To satisfy prohibitively massive computational requirements of current deep Convolutional Neural Networks (CNNs), CNN-specific accelerators are widely deployed in large-scale systems. Caused by high-energy neutrons and α-particle strikes, soft error may lead to catastrophic failures when CNN is deployed on high integration density accelerators. As CNNs become ubiquitous in mission-critical domains, ensuring the reliable execution of CNN accelerators in the presence of soft errors is increasingly essential. In this article, we propose to Re cycle I dle P rocessing E lements (PEs) in the CNN accelerator for vulnerable filters soft error detection (ReIPE). Considering the error-sensitivity of filters, ReIPE first carries out a filter-level gradient analysis process to replace fault injection for fast filter-wise error resilience estimation. Then, to achieve maximal reliability benefits, combining the hardware-level systolic array idleness and software-level CNN filter-wise error resilience profile, ReIPE preferentially duplicated loads the most vulnerable filters onto systolic array to recycle idle-column PEs for opportunistically redundant execution (error detection). Exploiting the data reuse properties of accelerators, ReIPE incorporates the error detection process into the original computation flow of accelerators to perform real-time error detection. Once the error is detected, ReIPE will trigger a correction round to rectify the erroneous output. Experimental results performed on LeNet-5, Cifar-10-CNN, AlexNet, ResNet-20, VGG-16, and ResNet-50 exhibit that ReIPE can cover 96.40% of errors while reducing 75.06% performance degradation and 67.79% energy consumption of baseline dual modular redundancy on average. Moreover, to satisfy the reliability requirements of various application scenarios, ReIPE is also applicable for pruned, quantized, and Transformer-based models, as well as portable to other accelerator architectures.