2016 IEEE Hot Chips 28 Symposium (HCS)最新文献_第2页

Passive dense stereo vision on the Myriad2 VPU 在Myriad2 VPU上的被动密集立体视觉

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936240

L. Puglia, M. Ionica, G. Raiconi, D. Moloney

引用次数: 4

VR and AR anytime and everywhere: Contributions of PMD depth sensing to an evolving ecosystem VR和AR随时随地:PMD深度传感对不断发展的生态系统的贡献

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936195

B. Buxbaum

引用次数: 0

3D reconstruction with tango 探戈三维重建

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936198

Ivan Dryanovski

引用次数: 1

KiloCore: A 32 nm 1000-processor array KiloCore:一个32nm的1000处理器阵列

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936218

Brent Bohnenstiehl, Aaron Stillmaker, J. Pimentel, Timothy Andreas, Bin Liu, A. Tran, E. Adeagbo, B. Baas

引用次数: 11

A 16nm 256-bit wide 89.6GByte/s total bandwidth in-package interconnect with 0.3V swing and 0.062pJ/bit power in InFO package 一个16nm 256位宽89.6GByte/s总带宽封装互连，0.3V摆幅和0.062pJ/bit功率在InFO封装

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936211

M. Lin, Chien-Chun Tsai, K. Hsieh, Wen-Hung Huang, Yu-Chi Chen, Shu-Chun Yang, Chin-Ming Fu, H. Zhan, Jinn-Yeh Chien, Shao-Yu Li, Y.-H. Chen, C. Kuo, S. Tai, Kazuyoshi Yamada

引用次数: 9

Deep compression and EIE: Efficient inference engine on compressed deep neural network 深度压缩与EIE:基于压缩深度神经网络的高效推理引擎

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936226

Song Han, Xingyu Liu, Huizi Mao, Jing Pu, A. Pedram, M. Horowitz, B. Dally

引用次数: 43

100Gbit/s, 120km, PAM 4 based switch to switch, layer 2 silicon photonics based optical interconnects for datacenters 100Gbit/s, 120km，基于PAM 4的交换机对交换机，用于数据中心的基于第2层硅光子学的光互连

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936212

R. Nagarajan, S. Bhoja, Tom Issenhuth

引用次数: 5

A new ×86 core architecture for the next generation of computing 下一代计算的新×86核心架构

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936224

Mike Clark

引用次数: 29

Memory technology and applications 存储技术及应用

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936173

Allen Rush

引用次数: 1

From model to FPGA: Software-hardware co-design for efficient neural network acceleration 从模型到FPGA:高效神经网络加速的软硬件协同设计

2016 IEEE Hot Chips 28 Symposium (HCS) Pub Date : 2016-08-01 DOI: 10.1109/HOTCHIPS.2016.7936208

Kaiyuan Guo, Lingzhi Sui, Jiantao Qiu, Song Yao, Song Han, Yu Wang, Huazhong Yang

{"title":"From model to FPGA: Software-hardware co-design for efficient neural network acceleration","authors":"Kaiyuan Guo, Lingzhi Sui, Jiantao Qiu, Song Yao, Song Han, Yu Wang, Huazhong Yang","doi":"10.1109/HOTCHIPS.2016.7936208","DOIUrl":"https://doi.org/10.1109/HOTCHIPS.2016.7936208","url":null,"abstract":"Artificial neural networks, which dominate artificial intelligence applications such as object recognition and speech recognition, are in evolution. To apply neural networks to wider applications, customized hardware are necessary since CPU and GPU are not efficient enough. FPGA can be an ideal platform for neural network acceleration since it is programmable and can achieve much higher energy efficiency compared with general-purpose processors. However, the long development period and insufficient performance of traditional FPGA acceleration solutions prevent it from wide utilization. In this work, we propose a complete design flow to achieve both fast deployment and high energy efficiency for accelerating neural networks on FPGA. Deep compression and data quantization are employed to exploit the redundancy in algorithm and reduce both computational and memory complexity. Two architecture designs for CNN and DNN/RNN will be introduced together with the compilation environment. Evaluated on Xilinx Zynq 7000 and Kintex Ultrascale series FPGA with realworld neural networks, up to 10 times higher energy efficiency can be achieved compared with mobile GPU and desktop GPU.","PeriodicalId":363333,"journal":{"name":"2016 IEEE Hot Chips 28 Symposium (HCS)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132296147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 42