DC Field | Value | Language |
---|---|---|
dc.contributor.author | Lee, Jaewon | ko |
dc.contributor.author | Kim, Gain | ko |
dc.contributor.author | Park, Jinho | ko |
dc.contributor.author | Bae, Hyeon-Min | ko |
dc.date.accessioned | 2021-11-02T06:48:09Z | - |
dc.date.available | 2021-11-02T06:48:09Z | - |
dc.date.created | 2021-10-26 | - |
dc.date.issued | 2021-05 | - |
dc.identifier.citation | IEEE International Symposium on Circuits and Systems (IEEE ISCAS) | - |
dc.identifier.issn | 0271-4302 | - |
dc.identifier.uri | http://hdl.handle.net/10203/288563 | - |
dc.description.abstract | In convolutional neural network (CNN) accelerators, the dominant power consumption is caused by the access of external data memory. In addition, power and area occupied by I/O interfaces maintaining low bit-error-rate, e.g., 1e-15, grow as the data rate increases. Considering the inherent error resilience of the inference process in machine learning applications, the requirement of error-free communication in the data-path is controversial. In this paper, a custom CNN accelerator integrating a channel emulator is designed by using an FPGA to analyze the effect of the BER of an I/O transceiver on the image classification accuracy. In order to implement a channel emulator, a digital-domain look-up-table (LUT)-based 12-tap FIR filter is employed to create inter-symbol interference (ISI), and a PRBS31 generator is used as a noise source. The implementation was evaluated by running the ImageNet dataset on the FPGA-based custom accelerator (Virtex Ultrascale+) implementing VGG-16. The results show that the BER up to 1e-4 in the memory access has a negligible impact on the inference accuracy. | - |
dc.language | English | - |
dc.publisher | IEEE | - |
dc.title | Link Bit-Error-Rate Requirement Analysis for Deep Neural Network Accelerators | - |
dc.type | Conference | - |
dc.identifier.wosid | 000696765400059 | - |
dc.identifier.scopusid | 2-s2.0-85109012400 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | IEEE International Symposium on Circuits and Systems (IEEE ISCAS) | - |
dc.identifier.conferencecountry | KO | - |
dc.identifier.conferencelocation | Daegu | - |
dc.identifier.doi | 10.1109/ISCAS51556.2021.9401112 | - |
dc.contributor.localauthor | Bae, Hyeon-Min | - |
dc.contributor.nonIdAuthor | Lee, Jaewon | - |
dc.contributor.nonIdAuthor | Kim, Gain | - |
dc.contributor.nonIdAuthor | Park, Jinho | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.