Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 4 of 4 for “"Winograd algorithm"”.

  1. Optimizing Convolutions in State-of-the-art Convolutional Neural Networks on Intel Xeon Phi

    … multiplication formulation, FFT formulation, Winograd algorithm, and direct convolution. Interestingly, most of these implementations target GPU architecture, which is mainly due to - 1) higher theoretical peak FLOPs of GPU and 2) popularity of CUDA deep learning library. However, not much …

    rice Repository record for Optimizing Convolutions in State-of-the-art Convolutional Neural Networks on Intel Xeon Phi (opens in a new tab)

  2. Optimization of Spatial Convolution in ConvNets on Intel KNL

    … here was to use an unconventional convolution algorithm called Winograd Convolution algorithm to decrease the number of arithmetic operations. However, the reduction of arithmetic operations in Winograd algorithm comes at the cost of complicating the memory accesses. So the objective was to use …

    ut-knoxville Repository record for Optimization of Spatial Convolution in ConvNets on Intel KNL (opens in a new tab)

  3. An Introduction to the Winograd Discrete Fourier Transform

    This paper illustrates Winograd's approach to computing the Discrete Fourier Transform (DFT). This new approach changes the DFT into a cyclic convolution of 2 sequences, and illustrates shortcuts for computing this cyclic convolution. This method is known to reduce the number of multiplies required …

    ucf Repository record for An Introduction to the Winograd Discrete Fourier Transform (opens in a new tab)