Summary of Vertical Symbolic Regression Via Deep Policy Gradient, by Nan Jiang et al.
Vertical Symbolic Regression via Deep Policy Gradient
by Nan Jiang, Md Nasim, Yexiang Xue
First submitted to arxiv on: 1 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 VSR, a novel approach for discovering symbolic equations with many independent variables from experimental data, has been successfully applied to various datasets. By reducing the search spaces following a vertical discovery path, VSR builds upon reduced-form equations involving a subset of independent variables to create full-fledged ones. While deep neural networks have shown great promise in scaling up VSR, combining them directly poses significant engineering challenges. To overcome these hurdles, we introduce Vertical Symbolic Regression using Deep Policy Gradient (VSR-DPG), which leverages a sequential decision-making process to build equations from repeated applications of grammar rules. Our approach outperforms popular baselines in identifying both algebraic and ordinary differential equations on various benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary VSR is a new way to find simple math formulas that use many variables from data. It works by starting with smaller formulas and building up to more complex ones. Deep learning, which is great at solving complex problems, can help make VSR better. However, combining these two approaches isn’t easy. To fix this, we created a new method called Vertical Symbolic Regression using Deep Policy Gradient (VSR-DPG). This method looks at finding math formulas as a series of small steps, where each step adds more complexity to the formula. We tested our approach on many datasets and found that it does better than other popular methods in identifying simple and complex math formulas. |
Keywords
* Artificial intelligence * Deep learning * Regression
