Summary of Supervised Batch Normalization, by Bilal Faye et al.
Supervised Batch Normalization
by Bilal Faye, Mustapha Lebbah, Hanane Azzag
First submitted to arxiv on: 27 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Supervised Batch Normalization (SBN) approach addresses the limitations of traditional Batch Normalization (BN) by expanding normalization beyond single mean and variance parameters, enabling the identification of data modes prior to training. This ensures effective normalization for samples sharing common features, categorized as contexts such as domains or modalities. The superiority of SBN is illustrated through various experiments on single and multi-task datasets, resulting in a 15.13% accuracy enhancement when integrated with Vision Transformer on CIFAR-100 and a 22.25% improvement on MNIST and SVHN in domain adaptation scenarios compared to BN. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary SBN is an innovative technique that helps neural networks learn better by normalizing data in a more effective way. Traditional Batch Normalization works well, but it can struggle when dealing with diverse datasets. SBN solves this problem by identifying patterns or “modes” in the data and adjusting its normalization accordingly. This leads to improved performance on various tasks and datasets. |
Keywords
» Artificial intelligence » Batch normalization » Domain adaptation » Multi task » Supervised » Vision transformer
