Summary of Efficient Model Compression Techniques with Fishleg, by Jamie Mcgowan et al.
Efficient Model Compression Techniques with FishLeg
by Jamie McGowan, Wei Sheng Lai, Weibin Chen, Henry Aldridge, Jools Clarke, Jezabel Garcia, Rui Xia, Yilei Liang, Guillaume Hennequin, Alberto Bernacchia
First submitted to arxiv on: 3 Dec 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 authors propose FishLeg surgeon, a novel second-order pruning method based on the Fisher-Legendre optimizer. This approach utilizes a meta-learning strategy to estimate the inverse Fisher information matrix, enabling flexible tensor factorization techniques that improve computational efficiency without sacrificing accuracy. The resulting method is less sensitive to stochasticity and allows for progressive refinement of estimates during gradual pruning. FishLeg achieves high or comparable performance with two common baselines in the area, including 84% accuracy at 95% sparsity on ResNet18 with CIFAR-10 data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FishLeg is a new way to make AI models smaller without losing their ability to work well. The authors want to help people with limited computer resources by creating a method that compresses neural networks while keeping them accurate. They use an optimization technique called Fisher-Legendre to estimate the importance of each weight and delete unimportant ones. This approach is better than previous methods because it’s more efficient, less sensitive to random errors, and can refine its estimates as it prunes the network. |
Keywords
» Artificial intelligence » Meta learning » Optimization » Pruning
