Full Approximation of Deep Neural Networks through Efficient Optimization.

Research output: Contribution to book/Conference proceedings/Anthology/ReportConference contributionContributedpeer-review

Contributors

Abstract

Approximate Computing is a promising paradigm for mitigating computational requirements of Deep Neural Networks (DNN), by taking advantage of their inherent error resilience. Specifically, the use of approximate multipliers in DNN inference can lead to significant improvements in power consumption of embedded DNN applications. This paper presents a methodology for efficient approximate multiplier selection and for full and uniform approximation of large DNNs, through retraining and minimization of the approximation error. We evaluate our methodology using 422 approximate multipliers from the EvoApprox library, with three different Residual architectures trained with Cifar10, and achieve energy savings of up to 18% surpassing the original floating-point accuracy, and of up to 58% with an accuracy loss of 0.73%.

Details

Original languageUndefined
Title of host publication2020 IEEE International Symposium on Circuits and Systems (ISCAS)
PublisherIEEE Xplore
Number of pages5
ISBN (print)978-1-7281-3320-1
Publication statusPublished - 2020
Peer-reviewedYes

Publication series

SeriesIEEE International Symposium on Circuits and Systems (ISCAS)
ISSN0271-4302

Keywords