CVLGJun 28, 2018

Differentiable Learning-to-Normalize via Switchable Normalization

arXiv:1806.10779v5191 citationsHas Code
Originality Incremental advance
AI Analysis

This addresses the challenge of normalization technique selection in deep learning, offering a flexible solution that is incremental over existing methods like group normalization.

The paper tackles the problem of selecting appropriate normalization methods for different layers in deep neural networks by proposing Switchable Normalization (SN), which learns to switch between channel, layer, and minibatch normalization, resulting in improved performance on benchmarks like ImageNet and COCO while maintaining robustness to small batch sizes (e.g., 2 images/GPU).

We address a learning-to-normalize problem by proposing Switchable Normalization (SN), which learns to select different normalizers for different normalization layers of a deep neural network. SN employs three distinct scopes to compute statistics (means and variances) including a channel, a layer, and a minibatch. SN switches between them by learning their importance weights in an end-to-end manner. It has several good properties. First, it adapts to various network architectures and tasks (see Fig.1). Second, it is robust to a wide range of batch sizes, maintaining high performance even when small minibatch is presented (e.g. 2 images/GPU). Third, SN does not have sensitive hyper-parameter, unlike group normalization that searches the number of groups as a hyper-parameter. Without bells and whistles, SN outperforms its counterparts on various challenging benchmarks, such as ImageNet, COCO, CityScapes, ADE20K, and Kinetics. Analyses of SN are also presented. We hope SN will help ease the usage and understand the normalization techniques in deep learning. The code of SN has been made available in https://github.com/switchablenorms/.

Code Implementations3 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes