In this work, we propose an effective scheme (called DP-Net) for compressing the deep neural networks (DNNs).
It is easy to train and fast to search.
Ranked #58 on Neural Architecture Search on ImageNet (MACs metric)
In this paper, we propose a novel meta learning approach for automatic channel pruning of very deep neural networks.
In this work, we propose a novel method called Meta-SR to firstly solve super-resolution of arbitrary scale factor (including non-integer scale factors) with a single model.