这里详述的核心概念很容易推广不同的模型。例如,直接运用于加速视频像素网络(https://arxiv.org/abs/1610.00527),并且由于更高的计算需求,可能会产生更好的加速效果。我们期待你对卷积自回归模型的快速生成的实际运用! 作者 Prajit Ramachandran (https://github.com/PrajitR) Tom Le Paine (https://github.com/tomlepaine) Pooya Khorrami (https://github.com/pkhorrami4) Mohammad Babaeizadeh (https://github.com/mbz) 如果你觉得这个项目有用,请引用我们已提交 ICLR 2017 的论文《FAST GENERATION FOR CONVOLUTIONAL AUTOREGRESSIVE MODELS》: @article{ramachandran2017fastgeneration, title={Fast Generation for Convolutional Autoregressive Models}, author={Ramachandran, Prajit and Paine, Tom Le and Khorrami, Pooya and Babaeizadeh, Mohammad and Chang, Shiyu and Zhang, Yang and Hasegawa-Johnson, Mark and Campbell, Roy and Huang, Thomas} year={2017}} ©本文为机器之心编译,转载请联系本公众号获得授权。 ?------------------------------------------------ 加入机器之心(全职记者/实习生):[email protected] 投稿或寻求报道:[email protected] 广告&商务合作:[email protected] ,atv直播 (责任编辑:本港台直播) |