Assessment of Machine Learning Algorithms for Near-Sensor Computing under Radiation Soft Errors

Machine learning (ML) algorithms have been regaining momentum thanks to their ability to analyze substantial and complex data, supporting artificial intelligence decisions in cloud computing but also in near-sensor computing in endpoint devices. Both cloud and near-sensor computing are liable to radiation-induced soft errors, especially in automotive and aerospace safety-critical applications. In this regard, this paper contributes by comparing the accuracy of two prominent machine learning algorithms running on a low-power processor upset by radiation-induced soft errors. Both ML algorithms have been assessed with the help of a fault injection-based method able to natively emulate soft errors directly in a development board. In addition, neutron radiation test results suggest the most critical situations in which mitigation solutions should address.

[1]  R. Fisher.  THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[2]  Luigi Carro,et al.  Analyzing and Increasing the Reliability of Convolutional Neural Networks on GPUs , 2019, IEEE Transactions on Reliability.

[3]  P. Rech,et al.  On the Reliability of Linear Regression and Pattern Recognition Feedforward Artificial Neural Networks in FPGAs , 2018, IEEE Transactions on Nuclear Science.

[4]  Ling Li,et al.  An Instruction Set Architecture for Machine Learning , 2019, ACM Trans. Comput. Syst..

[5]  Guanpeng Li,et al.  Understanding Error Propagation in Deep Learning Neural Network (DNN) Accelerators and Applications , 2017, SC17: International Conference for High Performance Computing, Networking, Storage and Analysis.

[6]  C. Frost,et al.  Selective Hardening for Neural Networks in FPGAs , 2019, IEEE Transactions on Nuclear Science.

[7]  Fadi J. Kurdahi,et al.  Are CNNs Reliable Enough for Critical Applications? An Exploratory Study , 2020, IEEE Design & Test.

[8]  Rodrigo Possamai Bastos,et al.  Assessment of a Hardware-Implemented Machine Learning Technique Under Neutron Irradiation , 2019, IEEE Transactions on Nuclear Science.

[9]  Gilles Sicard,et al.  Comparing transient-fault effects on synchronous and on asynchronous circuits , 2009, 2009 15th IEEE International On-Line Testing Symposium.

[10]  Nacer-Eddine Zergainoh,et al.  On the Robustness of Stochastic Bayesian Machines , 2017, IEEE Transactions on Nuclear Science.

[11]  Gu-Yeon Wei,et al.  Ares: A framework for quantifying the resilience of deep neural networks , 2018, 2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC).

[12]  Vikas Chandra,et al.  CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs , 2018, ArXiv.

[13]  Jeremy Kepner,et al.  Survey and Benchmarking of Machine Learning Accelerators , 2019, 2019 IEEE High Performance Extreme Computing Conference (HPEC).

[14]  Ricardo Reis,et al.  Using Machine Learning Techniques to Evaluate Multicore Soft Error Reliability , 2019, IEEE Transactions on Circuits and Systems I: Regular Papers.