Design of Analog and Digital Hybrid MAC Circuit for Artificial Neural Networks

被引：0

作者：

Park, Ki-Hyuk ^{[1
]}

Cho, Min-Hyung ^{[1
]}

Jeon, Young-Deuk ^{[1
]}

Lee, Joo-Hyun ^{[1
]}

机构：

[1] Elect & Telecommun Res Inst, Processor Res Grp, Daejeon, South Korea

来源：

2019 INTERNATIONAL CONFERENCE ON ELECTRONICS, INFORMATION, AND COMMUNICATION (ICEIC) | 2019年

关键词：

MAC; analog; digital; hybrid; neural network;

D O I：

10.23919/elinfocom.2019.8706340

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Demand for high-performance hardware acceleration for machine learning applications is increasing rapidly. This paper presents a low power analog and digital hybrid MAC (Multiply and Accumulation) circuit for artificial neural networks. The proposed MAC circuit consists of an analog synapse unit, digital preprocessing and postprocessing unit for support of the parallel analog synapse cores. As the hybrid MAC circuit supports relatively low power and fast multiple MAC operations, it provides a good advantage in developing hardware accelerator for artificial neural networks.

引用

页码：509 / 511

页数：3

共 50 条

[1] A multistable circuit for weight storage in analog artificial neural networks
Conti, M
Guaitini, G
Orcioni, S
Turchetti, C
[J]. ADVANCES IN INTELLIGENT SYSTEMS, 1997, 41 : 92 - 96
[2] CURRENT MODE ANALOG CIRCUIT FOR IMPLEMENTING ARTIFICIAL NEURAL NETWORKS
HOUSELANDER, PK
TAYLOR, JT
HAIGH, DG
[J]. ELECTRONICS LETTERS, 1988, 24 (10) : 630 - 631
[3] Design of An Analog Circuit-Based Artificial Neural Network
Gencer, Fikret Basar
Xhafa, Xhesila
Inam, Benan Beril
Yelten, Mustafa Berke
[J]. 2019 11TH INTERNATIONAL CONFERENCE ON ELECTRICAL AND ELECTRONICS ENGINEERING (ELECO 2019), 2019, : 379 - 383
[4] TIME-MULTIPLEXED ANALOG CIRCUIT FOR IMPLEMENTING ARTIFICIAL NEURAL NETWORKS
HOUSELANDER, PK
TAYLOR, JT
[J]. ELECTRONICS LETTERS, 1988, 24 (23) : 1413 - 1414
[5] ARTIFICIAL NEURAL NETWORKS AS AIDS IN-CIRCUIT DESIGN
KOTHAPALLI, G
[J]. MICROELECTRONICS JOURNAL, 1995, 26 (06) : 569 - 578
[6] An analog-digital hybrid CMOS circuit for two-dimensional motion detection with correlation neural networks
Asai, T
Koutani, M
Amemiya, Y
[J]. IJCNN 2000: PROCEEDINGS OF THE IEEE-INNS-ENNS INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOL III, 2000, : 494 - 499
[7] MIXED ANALOG-DIGITAL HARDWARE SYNTHESIS OF ARTIFICIAL NEURAL NETWORKS
ACHYUTHAN, A
ELMASRY, MI
[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 1994, 13 (09) : 1073 - 1087
[8] Digital design of sigmoid approximator for artificial neural networks
Basterretxea, K
Tarela, JM
del Campo, I
[J]. ELECTRONICS LETTERS, 2002, 38 (01) : 35 - 37
[9] Hybrid multiobjective evolutionary design for artificial neural networks
Goh, Chi-Keong
Teoh, Eu-Jin
Tan, Kay Chen
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2008, 19 (09): : 1531 - 1548
[10] Hybrid heuristics for optimal design of Artificial Neural Networks
Abraham, A
Nath, B
[J]. DEVELOPMENTS IN SOFT COMPUTING, 2001, : 15 - 22

← 1 2 3 4 5 →