Litcius/Paper detail

MOZART+: Masking Outputs With Zeros for Improved Architectural Robustness and Testing of DNN Accelerators

Stéphane Burel, Adrian Evans, Lorena Anghel

2022IEEE Transactions on Device and Materials Reliability28 citationsDOIOpen Access PDF

Abstract

Deep Neural Networks (DNNs) are increasingly used in safety critical autonomous systems. We present MOZART+, a DNN accelerator architecture which provides fault detection and fault tolerance. MOZART+ is a systolic architecture based on the Output Stationary (OS) data-flow, as it is a data-flow that inherently limits fault propagation. In addition, MOZART+ achieves fault detection with on-line functional testing of the Processing Elements (PEs). Faulty PEs are swiftly taken off-line with minimal classification impact. We show how to handle the case of layers with a small number of neurons. The implementation of our approach on Squeezenet results in a loss of accuracy of less than 3% in the presence of a single faulty PE, compared to 15-33% without mitigation. The area overhead for the test logic does not exceed 8%. Dropout during training further improves fault tolerance, without a priori knowledge of the faults. We present a detailed fault-injection study on multiple systolic architectures, considering different fault-models and comparing different measures of accuracy.

Topics & Concepts

Robustness (evolution)Computer scienceFault toleranceOverhead (engineering)MOZARTArtificial neural networkFault detection and isolationA priori and a posterioriFault coverageComputer engineeringReal-time computingReliability engineeringEmbedded systemArtificial intelligenceEngineeringDistributed computingElectronic circuitElectrical engineeringBiochemistryActuatorArt historyPhilosophyOperating systemGeneEpistemologyArtChemistryAdversarial Robustness in Machine LearningRadiation Effects in ElectronicsAdvanced Neural Network Applications
MOZART+: Masking Outputs With Zeros for Improved Architectural Robustness and Testing of DNN Accelerators | Litcius