| Title | MILR: Mathematically Induced Layer Recovery for Plaintext Space Error Correction of CNNs |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Ponader, Jonathan, Thomas, Kyle, Kundu, Sandip, Solihin, Yan |
| Conference Name | 2021 51st Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN) |
| Date Published | jun |
| Keywords | convolutional neural networks, Elliptic curve cryptography, Encryption, error correction, error detection, error recovery, Hardware, Mission critical systems, Neural Network, neural network resiliency, pubcrawl, resilience, Resiliency, Robustness, Software |
| Abstract | The increased use of Convolutional Neural Networks (CNN) in mission-critical systems has increased the need for robust and resilient networks in the face of both naturally occurring faults as well as security attacks. The lack of robustness and resiliency can lead to unreliable inference results. Current methods that address CNN robustness require hardware modification, network modification, or network duplication. This paper proposes MILR a software-based CNN error detection and error correction system that enables recovery from single and multi-bit errors. The recovery capabilities are based on mathematical relationships between the inputs, outputs, and parameters(weights) of the layers; exploiting these relationships allows the recovery of erroneous parameters (iveights) throughout a layer and the network. MILR is suitable for plaintext-space error correction (PSEC) given its ability to correct whole-weight and even whole-layer errors in CNNs. |
| DOI | 10.1109/DSN48987.2021.00024 |
| Citation Key | ponader_milr_2021 |
MILR: Mathematically Induced Layer Recovery for Plaintext Space Error Correction of CNNs