The Indirect Convolution Algorithm
arXiv:1907.02129
Abstract
Deep learning frameworks commonly implement convolution operators with GEMM-based algorithms. In these algorithms, convolution is implemented on top of matrix-matrix multiplication (GEMM) functions, provided by highly optimized BLAS libraries. Convolutions with 1x1 kernels can be directly represented as a GEMM call, but convolutions with larger kernels require a special memory layout transformation - im2col or im2row - to fit into GEMM interface. The Indirect Convolution algorithm provides the efficiency of the GEMM primitive without the overhead of im2col transformation. In contrast to GEMM-based algorithms, the Indirect Convolution does not reshuffle the data to fit into the GEMM primitive but introduces an indirection buffer - a buffer of pointers to the start of each row of image pixels. This broadens the application of our modified GEMM function to convolutions with arbitrary kernel size, padding, stride, and dilation. The Indirect Convolution algorithm reduces memory overhead proportionally to the number of input channels and outperforms the GEMM-based algorithm by up to 62% on convolution parameters which involve im2col transformations in GEMM-based algorithms. This, however, comes at cost of minor performance reduction on 1x1 stride-1 convolutions.
Presented on Efficient Deep Learning for Computer Vision workshop at CVPR 2019
References in corpus (8)
- YOLOv3: An Incremental Improvement
- WaveNet: A Generative Model for Raw Audio
- cuDNN: Efficient Primitives for Deep Learning
- Fast Convolutional Nets With fbfft: A GPU Performance Evaluation
- Panoptic Feature Pyramid Networks
- Low-memory GEMM-based convolution algorithms for deep neural networks
- High Performance Zero-Memory Overhead Direct Convolutions
- Pooling Pyramid Network for Object Detection
Cited by in corpus (4)
- Photonic Convolution Neural Network Based on Interleaved Time-Wavelength Modulation
- TCN Mapping Optimization for Ultra-Low Power Time-Series Edge Inference
- Automated Design Space Exploration for optimised Deployment of DNN on Arm Cortex-A CPUs
- High Performance and Portable Convolution Operators for ARM-based Multicore Processors