arxivst stuff from arxiv that you should probably bookmark

Fast Generation for Convolutional Autoregressive Models

Abstract · Apr 20, 2017 04:13 ·

autoregressive generation wavenet fast models den 2016a oord cs-lg cs-cv stat-ml

Arxiv Abstract

  • Prajit Ramachandran
  • Tom Le Paine
  • Pooya Khorrami
  • Mohammad Babaeizadeh
  • Shiyu Chang
  • Yang Zhang
  • Mark A. Hasegawa-Johnson
  • Roy H. Campbell
  • Thomas S. Huang

Convolutional autoregressive models have recently demonstrated state-of-the-art performance on a number of generation tasks. While fast, parallel training methods have been crucial for their success, generation is typically implemented in a na\“{i}ve fashion where redundant computations are unnecessarily repeated. This results in slow generation, making such models infeasible for production environments. In this work, we describe a method to speed up generation in convolutional autoregressive models. The key idea is to cache hidden states to avoid redundant computation. We apply our fast generation method to the Wavenet and PixelCNN++ models and achieve up to $21\times$ and $183\times$ speedups respectively.

Read the paper (pdf) »