Publications

Detailed Information

Extreme Partial-Sum Quantization for Analog Computing-In-Memory Neural Network Accelerators

DC Field Value Language
dc.contributor.authorKim, Yulhwa-
dc.contributor.authorKim, Hyungjun-
dc.contributor.authorKim, Jae-Joon-
dc.date.accessioned2023-01-09T02:50:04Z-
dc.date.available2023-01-09T02:50:04Z-
dc.date.created2022-12-14-
dc.date.issued2022-10-
dc.identifier.citationACM Journal on Emerging Technologies in Computing Systems, Vol.18 No.4, p. 75-
dc.identifier.issn1550-4832-
dc.identifier.urihttps://hdl.handle.net/10371/188964-
dc.description.abstractIn Analog Computing-in-Memory (CIM) neural network accelerators, analog-to-digital converters (ADCs) are required to convert the analog partial sums generated from a CIM array to digital values. The overhead from ADCs substantially degrades the energy efficiency of CIM accelerators so that previous works attempted to lower the ADC resolution considering the distribution of the partial sums. Despite the efforts, the required ADC resolution still remains relatively high. In this article, we propose the data-driven partial sum quantization scheme, which exhaustively searches for the optimal quantization range with little computational burden. We also report that analyzing the characteristics of the partial sum distributions at each layer gives an additional information to further reduce the ADC resolution compared to previous works that mostly used the characteristics of the partial sum distributions of the entire network. Based on the finer-level data-driven approach combined with retraining, we present a methodology for extreme partial-sum quantization. Experimental results show that the proposed method can reduce the ADC resolution to 2 to 3 bits for CIFAR-10 dataset, which is the smaller ADC bit resolution than any previous CIM-based NN accelerators.-
dc.language영어-
dc.publisherAssociation for Computing Machinary, Inc.-
dc.titleExtreme Partial-Sum Quantization for Analog Computing-In-Memory Neural Network Accelerators-
dc.typeArticle-
dc.identifier.doi10.1145/3528104-
dc.citation.journaltitleACM Journal on Emerging Technologies in Computing Systems-
dc.identifier.wosid000885840300013-
dc.citation.number4-
dc.citation.startpage75-
dc.citation.volume18-
dc.description.isOpenAccessN-
dc.contributor.affiliatedAuthorKim, Jae-Joon-
dc.type.docTypeArticle-
dc.description.journalClass1-
Appears in Collections:
Files in This Item:
There are no files associated with this item.

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share