Fault-Tolerant Dot-Product Engines

2017/08/23 by Roth, Ron M. · 1 citation
#FOS: Computer and information sciences #Information Theory (cs.IT)

paper · doi:10.48550/arxiv.1708.06892

Abstract

Coding schemes are presented that provide the ability to correct and detect computational errors while using dot-product engines for integer vector--matrix multiplication. Both the L₁-metric and the Hamming metric are considered.

Cited by

• Low-Rank-Based Approximate Computation with Memristors

Related

• Error correction in fast matrix multiplication and inverse
• Rank error-correcting pairs
• Linear Computation Coding
• Grid-like Error-Correcting Codes for Matrix Multiplication with Better Correcting Capability
• High-Performance Level-1 and Level-2 BLAS