Summary of Shapley Pruning For Neural Network Compression, by Kamil Adamczewski et al.
Shapley Pruning for Neural Network Compression
by Kamil Adamczewski, Yawei Li, Luc van Gool
First submitted to arxiv on: 19 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
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 Medium Difficulty summary: This paper proposes a novel framework for neural network pruning, combining existing methods like leave-one-out pruning and oracle pruning into a Shapley value-based approach. The goal is to compress convolutional neural networks while maintaining performance. To make this practical, the authors provide approximations of the Shapley value and evaluate their ranking method against a new benchmark, Oracle rank, based on oracle sets. Experimental results show that the proposed approach obtains state-of-the-art network compression. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty summary: This research paper is about making neural networks smaller without losing their ability to do tasks well. Neural networks are like big teams working together, and this paper shows how to pick which team members are most important and remove the rest. The goal is to make these networks use less computer power and energy while still doing a good job. The scientists developed a new way to figure out which parts of the network are most important and tested it with real-world data. |
Keywords
» Artificial intelligence » Neural network » Pruning
