Please use this identifier to cite or link to this item:
|Title:||Analysis of circuit aging on accuracy degradation of deep neural network accelerator||Authors:||Liu, Wenye
|Keywords:||Engineering::Electrical and electronic engineering::Integrated circuits||Issue Date:||2019||Source:||Liu, W. & Chang, C.-H. (2019). Analysis of circuit aging on accuracy degradation of deep neural network accelerator. 2019 IEEE International Symposium on Circuits and Systems (ISCAS). doi:10.1109/ISCAS.2019.8702226||Project:||MOE2015-T2-2-013||Abstract:||Deep neural networks have achieved phenomenal successes in vision recognition tasks, which motivate the deployment of deep learning in portable and smart wearable devices. To overcome the fundamental challenges of power and resource limitation, application-specific integrated circuit accelerators have emerged to compact the model and use lower precision arithmetic to increase the throughput of computation with reduced power consumption. Although very high energy efficiency has been achieved by removing redundant weights, compressing data and even sacrificing timing margin, such trend in hardware acceleration that pushes the deep learning systems to the error threshold can be disastrous for the tasks they performed due to failure or degraded performance of circuit components. Concerned by the lack of attention on the evolving unreliability effects in artificial intelligent accelerators implemented by the continuously scaled CMOS technology, this paper is the first to evaluate the effect of circuit aging on performance degradation of deep learning accelerator. Our findings indicate that DNN system running at their peak throughput rate can experience up to 84% accuracy drop after a year of aging and the accumulation of errors aggravates with the depth of learning. It is also found that relaxation of throughput rate can slow down the loss of classification accuracy considerably.||URI:||https://hdl.handle.net/10356/136843||ISBN:||9781728103976||DOI:||10.1109/ISCAS.2019.8702226||Rights:||© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/ISCAS.2019.8702226||Fulltext Permission:||open||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Conference Papers|
Updated on Apr 10, 2021
Updated on Apr 10, 2021
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.