Precision-Energy-Throughput Scaling Of Generic Matrix Multiplication and Convolution Kernels Via Linear Projections

TL;DR

This paper introduces a linear projection-based method to scale energy and throughput in GEMM and CONV kernels for multimedia applications, achieving significant efficiency gains with minimal accuracy loss.

Contribution

It presents a novel approach using linear projections to approximate GEMM and CONV computations, enabling scalable energy and throughput improvements in error-tolerant multimedia tasks.

Findings

Achieves 280%-440% throughput increase

Reduces energy consumption by 75%-80%

Maintains accuracy in face recognition and music matching

Abstract

Generic matrix multiplication (GEMM) and one-dimensional convolution/cross-correlation (CONV) kernels often constitute the bulk of the compute- and memory-intensive processing within image/audio recognition and matching systems. We propose a novel method to scale the energy and processing throughput of GEMM and CONV kernels for such error-tolerant multimedia applications by adjusting the precision of computation. Our technique employs linear projections to the input matrix or signal data during the top-level GEMM and CONV blocking and reordering. The GEMM and CONV kernel processing then uses the projected inputs and the results are accumulated to form the final outputs. Throughput and energy scaling takes place by changing the number of projections computed by each kernel, which in turn produces approximate results, i.e. changes the precision of the performed computation. Results derived from a voltage- and frequency-scaled ARM Cortex A15 processor running face recognition and music matching algorithms demonstrate that the proposed approach allows for 280%~440% increase of processing throughput and 75%~80% decrease of energy consumption against optimized GEMM and CONV kernels without any impact in the obtained recognition or matching accuracy. Even higher gains can be obtained if one is willing to tolerate some reduction in the accuracy of the recognition and matching applications.