Summary of Sampa: Sharpness-aware Minimization Parallelized, by Wanyun Xie et al.
SAMPa: Sharpness-aware Minimization Parallelized
by Wanyun Xie, Thomas Pethick, Volkan Cevher
First submitted to arxiv on: 14 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 SAMPa method is a modification of Sharpness-aware minimization (SAM) that enables parallel computation of gradients, achieving a twofold speedup over SAM. This improvement maintains convergence guarantees for fixed perturbation sizes, as demonstrated by a novel Lyapunov function. Empirical results show that SAMPa outperforms SAM in both vision and language tasks, with efficient computational time. The code is available at https://github.com/LIONS-EPFL/SAMPa. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary SAMPa is a way to make neural networks better by improving how they’re trained. It’s based on another method called SAM, but SAMPa makes the training process faster and more efficient. This means that it can learn from data faster and be used for even more tasks. The new method also works well with different types of data, such as pictures or text. Having this code available will help researchers use SAMPa to improve their own projects. |
Keywords
» Artificial intelligence » Sam
