Summary of Symbolic Regression Via Mdlformer-guided Search: From Minimizing Prediction Error to Minimizing Description Length, by Zihan Yu et al.
Symbolic regression via MDLformer-guided search: from minimizing prediction error to minimizing description length
by Zihan Yu, Jingtao Ding, Yong Li
First submitted to arxiv on: 6 Nov 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 novel symbolic regression method, SR4MDL, proposed in this paper addresses the limitations of existing methods by introducing a minimum description length-based search objective. This approach leverages a neural network, MDLformer, to estimate the distance from the target formula, enabling robust and scalable estimation. By using the MDLformer’s output as the search objective, SR4MDL effectively recovers the correct mathematical form of formulas from data, outperforming state-of-the-art methods on two benchmark datasets. The method is further demonstrated to generalize well to unseen black-box problems. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a new way to find the formula that best fits some given data. It uses a type of neural network called MDLformer to estimate how close it is to finding the correct formula, and then uses this information to guide its search for the best formula. The method, called SR4MDL, is able to recover the correct formula from data much better than existing methods can. |
Keywords
» Artificial intelligence » Neural network » Regression
