Gene

Quantum mechanics

Algorithm

Soft error

Engineering

Robustness (evolution)

Computation

Bridge (graph theory)

Telecommunications

Power (physics)

Set (abstract data type)

Biochemistry

Parallel computing

Computer science

Distributed computing

Software

Speedup

Physics

Fault injection

Fault tolerance

Ranging

Electronic engineering

Computer engineering

Internal medicine

Reliability (semiconductor)

Artificial intelligence

Medicine

Chemistry

Programming language

Error detection and correction

Convolutional neural network

Propagation of uncertainty

IEEE Transactions on Computers

The great quest for adopting AI-based computation for safety-/mission-critical applications motivates the interest towards methods for assessing the robustness of the application w.r.t. not only its training/tuning but also errors due to faults, in particular soft errors, affecting the underlying hardware. Two strategies exist: architecture-level fault injection and application-level functional error simulation. We present a framework for the reliability analysis of Convolutional Neural Networks (CNNs) via an error simulation engine that exploits a set of validated error models extracted from a detailed fault injection campaign. These error models are defined based on the corruption patterns of the output of the CNN operators induced by faults and bridge the gap between fault injection and error simulation, exploiting the advantages of both approaches. We compared our methodology against SASSIFI for the accuracy of functional error simulation w.r.t. fault injection, and against TensorFI in terms of speedup for the error simulation strategy. Experimental results show that our methodology achieves about 99\% accuracy of the fault effects w.r.t. SASSIFI, and a speedup ranging from 44x up to 63x w.r.t. TensorFI, that only implements a limited set of error models.

Fast and Accurate Error Simulation for CNNs Against Soft Errors