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。

ISSCC 2017 Session14 Deep Learning Processors:

ISSCC 2017关于Deep Learning Processors的Slides笔记，主要参考了[1]中的笔记，自己根据paper和slides读一遍，这里记一下笔记，方便以后查阅。

14.1 A 2.9TOPS/W Deep Convolutional Neural Network SoC in FD-SOI 28nm for Intelligent Embedded Systems

14.2 DNPU: An 8.1TOPS/W Reconfigurable CNN-RNN Processor for General-Purpose Deep Neural Networks

14.3 A 28nm SoC with a 1.2GHz 568nJ/Prediction Sparse Deep-Neural-Network Engine with >0.1 Timing Error Rate Tolerance for IoT Applications

14.4 A Scalable Speech Recognizer with Deep-Neural-Network Acoustic Models and Voice-Activated Power Gating

14.5 ENVISION: A 0.26-to-10TOPS/W Subword-Parallel Dynamic-Voltage-Accuracy-Frequency-Scalable Convolutional Neural Network Processor in 28nm FDSOI

14.6 A 0.62mW Ultra-Low-Power Convolutional-Neural-Network Face-Recognition Processor and a CIS Integrated with Always-On Haar-Like Face Detector

14.7 A 288μW Programmable Deep-Learning Processor with 270KB On-Chip Weight Storage Using Non-Uniform Memory Hierarchy for Mobile Intelligence

14.8 A 135mW Fully Integrated Data Processor for Next-Generation Sequencing

ASPLOS'17论文导读——SC-DCNN: Highly-Scalable Deep Convolutional Neural Network using Stochastic Computing

FPGA 17最佳论文导读 ESE: Efficient Speech Recognition Engine with Compressed LSTM on FPGA

ISCA 2017：

Machine Learning Session 1,2：

待补充...

参考资料

[1] https://reconfigdeeplearning.com/2017/02/10/isscc-2017-session-14-dlp/