Positional Normalization
This work addresses the training efficiency and structural preservation in generative networks, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of reducing training time in deep neural networks by proposing a new normalization method that normalizes exclusively across channels instead of spatial dimensions, allowing the extraction and reuse of structural information from normalization constants in generative networks.
A popular method to reduce the training time of deep neural networks is to normalize activations at each layer. Although various normalization schemes have been proposed, they all follow a common theme: normalize across spatial dimensions and discard the extracted statistics. In this paper, we propose an alternative normalization method that noticeably departs from this convention and normalizes exclusively across channels. We argue that the channel dimension is naturally appealing as it allows us to extract the first and second moments of features extracted at a particular image position. These moments capture structural information about the input image and extracted features, which opens a new avenue along which a network can benefit from feature normalization: Instead of disregarding the normalization constants, we propose to re-inject them into later layers to preserve or transfer structural information in generative networks. Codes are available at https://github.com/Boyiliee/PONO.