#ISSCC

ISSCC 2017论文导读 Session 14:A 288μW Programmable Deep-Learning Processor with 270KB On-Chip Weight

A288μWProgrammableDeep-LearningProcessorwith270KBOn-ChipWeightStorageUsingNon-UniformMemoryHierarchyforMobileIntelligence单位:Michigan,CubeWorks(密歇根大学,CubeWorks公司...

[置顶] 人工智能(深度学习)加速芯片论文阅读笔记 (已添加ISSCC17,FPGA17...ISCA17...)

这是一个导读,可以快速找到我记录的关于人工智能(深度学习)加速芯片论文阅读笔记。ISSCC2017Session14DeepLearningProcessors:ISSCC2017关于DeepLearningProcessors的Slides笔记,主要参考了[1]中的笔记,自己根据paper和slides读一遍,这里记...

ISSCC 2017论文导读 Session 14:A 0.62mW Ultra-Low-Power Convolutional-Neural-Network Face-Recognition Pro

A0.62mWUltra-Low-PowerConvolutional-Neural-NetworkFace-RecognitionProcessorandaCISIntegratedwithAlways-OnHaar-LikeFaceDetector单位:KAIST(韩国科学技术院)——ISSCC上大神级的机构···...

ISSCC 2017论文导读 Session 14:ENVISION: A 0.26-to-10 TOPS/W Subword-Parallel DVAFS CNN Processor in 28nm

ENVISION:A0.26-to-10TOPS/WSubword-ParallelDynamic-Voltage-Accuracy-Frequency-ScalableCNNProcessorin28nmFDSOI单位:EAST-MICAS,KULeuven(鲁汶大学)本文是我觉得本次ISSCC2017session...

ISSCC 2017论文导读 Session 14: A 28nm SoC with a 1.2GHz Prediction Sparse Deep-Neural-Network Engine

A28nmSoCwitha1.2GHz568nJ/PredictionSparseDeep-Neural-NetworkEnginewith>0.1TimingErrorRateToleranceforIoTApplications单位:Harvard(哈佛大学)这是一篇专门为DNN加速设计的芯片,在CNN加速芯...

ISSCC 2017论文导读 Session 14 Deep Learning Processors,DNPU: An 8.1TOPS/W Reconfigurable CNN-RNN

转载请注明,本文出自Bin的专栏http://blog.csdn.net/xbinworld,谢谢!DNPU:An8.1TOPS/WReconfigurableCNN-RNNProcessorforGeneral-PurposeDeepNeuralNetworks单位:KAIST(韩国科学技术院,电子工程-半导体系统实...

ISSCC 2017论文导读 Session 14 Deep Learning Processors,A 2.9TOPS/W Deep Convolutional Neural Network

最近ISSCC2017大会刚刚举行,看了关于DeepLearning处理器的Session14,有一些不错的东西,在这里记录一下。A2.9TOPS/WDeepConvolutionalNeuralNetworkSoCinFD-SOI28nmforIntelligentEmbeddedSystems单位:STMicroe...