人工神经网络在FPGA中的实现:一个案例研究

2016 International SoC Design Conference (ISOCC) Pub Date : 2016-10-01 DOI:10.1109/ISOCC.2016.7799795

Shuai Li, K. Choi, Yunsik Lee

{"title":"人工神经网络在FPGA中的实现:一个案例研究","authors":"Shuai Li, K. Choi, Yunsik Lee","doi":"10.1109/ISOCC.2016.7799795","DOIUrl":null,"url":null,"abstract":"Artificial Neural Network (ANN) is very powerful to deal with signal processing, computer vision and many other recognition problems. In this work, we implement basic ANN in FPGA. Compared with software, the FPGA implementation can utilize parallelism to speedup processing time. Additionally, hardware implementation can save more power compared with CPU/GPU. Our ANN in FPGA has a high learning ability, for logical XOR problem, which reduced the error rate from 10-2 to 10-4.","PeriodicalId":278207,"journal":{"name":"2016 International SoC Design Conference (ISOCC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":"{\"title\":\"Artificial neural network implementation in FPGA: A case study\",\"authors\":\"Shuai Li, K. Choi, Yunsik Lee\",\"doi\":\"10.1109/ISOCC.2016.7799795\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Artificial Neural Network (ANN) is very powerful to deal with signal processing, computer vision and many other recognition problems. In this work, we implement basic ANN in FPGA. Compared with software, the FPGA implementation can utilize parallelism to speedup processing time. Additionally, hardware implementation can save more power compared with CPU/GPU. Our ANN in FPGA has a high learning ability, for logical XOR problem, which reduced the error rate from 10-2 to 10-4.\",\"PeriodicalId\":278207,\"journal\":{\"name\":\"2016 International SoC Design Conference (ISOCC)\",\"volume\":\"15 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"9\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 International SoC Design Conference (ISOCC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISOCC.2016.7799795\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 International SoC Design Conference (ISOCC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISOCC.2016.7799795","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 9

摘要

人工神经网络(ANN)在处理信号处理、计算机视觉和许多其他识别问题方面非常强大。在本工作中，我们在FPGA上实现了基本的人工神经网络。与软件相比，FPGA实现可以利用并行性来加快处理时间。此外，与CPU/GPU相比，硬件实现可以节省更多的功耗。我们在FPGA中的人工神经网络具有很高的学习能力，对于逻辑异或问题，将错误率从10-2降低到10-4。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Artificial neural network implementation in FPGA: A case study

Artificial Neural Network (ANN) is very powerful to deal with signal processing, computer vision and many other recognition problems. In this work, we implement basic ANN in FPGA. Compared with software, the FPGA implementation can utilize parallelism to speedup processing time. Additionally, hardware implementation can save more power compared with CPU/GPU. Our ANN in FPGA has a high learning ability, for logical XOR problem, which reduced the error rate from 10-2 to 10-4.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2016 International SoC Design Conference (ISOCC)

自引率

0.00%

发文量