Summary of Order Of Compression: a Systematic and Optimal Sequence to Combinationally Compress Cnn, by Yingtao Shen et al.
Order of Compression: A Systematic and Optimal Sequence to Combinationally Compress CNN
by Yingtao Shen, Minqing Sun, Jianzhe Lin, Jie Zhao, An Zou
First submitted to arxiv on: 26 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); Neural and Evolutionary Computing (cs.NE)
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 A novel approach to neural network compression is proposed, addressing the issue of combining different compression techniques in an optimal manner. The authors introduce the Order of Compression, a systematic method for sequencing multiple compression approaches to achieve maximum effectiveness. This technique leverages topological sorting to determine the most efficient order, which is then validated on image-based regression and classification networks across various datasets. The results demonstrate significant reductions in computational costs, up to 859 times on ResNet34, with minimal accuracy loss (-0.09% for CIFAR10), compared to the baseline model. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Neural network compression helps make machine learning models smaller and faster. This paper shows that combining different ways of compressing neural networks is tricky because the order matters. The authors create a system called Order of Compression, which finds the best order to apply multiple compression techniques. They test this on image recognition and classification tasks and show that it can speed up computations by up to 859 times without losing much accuracy. |
Keywords
* Artificial intelligence * Classification * Machine learning * Neural network * Regression
