Convolution

• Convolution in Caffe: a memo

• How does 3D convolution really works in CAFFE, a detailed analysis

• Why GEMM is at the heart of deep learning

• Convolution: understand the mathematics

convolution types

• Advanced convolution techniques and layers

• Different types of the convolution layers

• A Comprehensive Introduction to Different Types of Convolutions in Deep Learning

• Talented Mr. 1X1: Comprehensive look at 1X1 Convolution in Deep Learning

• An Illustrated Explanation of Performing 2D Convolutions Using Matrix Multiplications

• An Introduction to different Types of Convolutions in Deep Learning

depthwise convolution

• 为什么depthwise convolution 比 convolution更加耗时？

atrous convolution

• 总结-空洞卷积(Dilated/Atrous Convolution)
• atrous convolution是什么？
• Atrous Convolution

Winograd

• https://www.encyclopediaofmath.org/index.php/Winograd_small_convolution_algorithm

• 深度学习（十一）——Winograd（1）

• 深度学习（十一）——Winograd（1）

• 深度学习（十二）——Winograd（2）

• Winograd大白话

• "Not so fast, FFT": Winograd

• Going beyond full utilization: The inside scoop on Nervana's Winograd kernels

• 源于《孙子算经》的Cudnn

• 黄教主的开源NVDLA，“核”爆还是“核”平？

• [DL]Winograd快速卷积算法

• https://github.com/chasingegg/Winconv

FFT

• The fastest convolutions in Theano with meta-optimization

• CUDA并行算法系列之FFT快速卷积

• https://github.com/chrischoy/CUDA-FFT-Convolution

• https://github.com/vdumoulin/conv_arithmetic

Implementation

RenderScript

• https://developer.android.com/reference/android/renderscript/ScriptIntrinsicConvolve3x3.html
• https://developer.android.com/reference/android/renderscript/ScriptIntrinsicConvolve5x5.html
• https://github.com/mtmd/Mobile_ConvNet

OpenCL

• Optimizing Mobile Deep Learning on ARM GPU with TVM

MXNet

• Operators in MXNet-Convolution

Snapdragon Neural Processing Engine

• https://developer.qualcomm.com/software/snapdragon-neural-processing-engine-ai
• 安装使用Qualcomm® Snapdragon™ Neural Processing Engine (NPE) SDK

OpenBLAS

• OpenBLAS矩阵乘法源码结构分析