Summary of Sharpness-aware Minimization For Evolutionary Feature Construction in Regression, by Hengzhe Zhang et al.
Sharpness-Aware Minimization for Evolutionary Feature Construction in Regression
by Hengzhe Zhang, Qi Chen, Bing Xue, Wolfgang Banzhaf, Mengjie Zhang
First submitted to arxiv on: 11 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 This research proposes a novel approach to genetic programming (GP)-based evolutionary feature construction that mitigates overfitting. By drawing inspiration from PAC-Bayesian theory and incorporating sharpness-aware minimization in function space, the method discovers symbolic features with robust performance within a smooth loss landscape. The proposed method optimizes sharpness along with cross-validation loss, and incorporates a sharpness reduction layer to control overfitting. Experimental results on 58 real-world regression datasets demonstrate that the approach outperforms standard GP as well as six state-of-the-art complexity measurement methods for GP. Additionally, the ensemble version of GP with sharpness-aware minimization shows superior performance compared to nine fine-tuned machine learning and symbolic regression algorithms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses a new way to build features using genetic programming (GP) that helps prevent overfitting. Overfitting is when a model becomes too specialized in its training data and doesn’t work well on new, unseen data. The researchers took ideas from a field called PAC-Bayesian theory and applied them to GP. They also added two new parts: sharpness-aware minimization and a sharpness reduction layer. This helped the method perform better when there wasn’t much training data or when the labels were noisy. The results showed that this approach worked well on 58 real-world datasets, beating standard GP and other methods. |
Keywords
» Artificial intelligence » Machine learning » Overfitting » Regression
