Quantized neural networks (QNNs) are increasingly considered for adoption on multiple applications, thanks to their high accuracy, but also since they allow for significantly lower compute and memory footprints. While the theory behind QNNs is achieving a high level of maturity, several new challenges have arisen during QNN deployment. Reliable and safe implementations of QNN accelerators becomes pivotal, especially when targeting safety critical applications like automotive, industrial and aerospace, requiring innovative solutions and their careful evaluation. In this work we compare the accuracy of QNNs during accelerated radiation testing when trained with different methodologies and implemented with a dataflow architecture in field programmable gate arrays (FPGA). The initial experiment shows that QNNs trained with a novel methodology, called fault-aware training (FAT), which accounts for soft errors during neural network (NN) training, makes QNNs more resilient to single-event-effects (SEEs) in FPGA.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Accelerated Radiation Test on Quantized Neural Networks trained with Fault Aware Training


    Contributors:


    Publication date :

    2022-03-05


    Size :

    1576611 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Channel-Aware Tracking in Multi-Hop Wireless Sensor Networks with Quantized Measurements

    Xiaojun Yang, / Ruixin Niu, / Masazade, Engin et al. | IEEE | 2013


    Neural networks robot controller trained with evolution strategies

    Berlanga de Jesús, Antonio / Isasi, Pedro / Sanchis de Miguel, María Araceli et al. | BASE | 1999

    Free access


    Traffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks

    Jin, Junqi / Fu, Kun / Zhang, Changshui | IEEE | 2014