Summary of Mmsr: Symbolic Regression Is a Multi-modal Information Fusion Task, by Yanjie Li et al.
MMSR: Symbolic Regression is a Multi-Modal Information Fusion Task
by Yanjie Li, Jingyi Liu, Weijun Li, Lina Yu, Min Wu, Wenqiang Li, Meilan Hao, Su Wei, Yusong Deng
First submitted to arxiv on: 28 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
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 Multi-Modal Symbolic Regression (MMSR) model tackles the long-standing challenge of symbolic regression, which seeks to describe complex laws of nature with concise mathematical formulas. Building upon previous work in Genetic Programming and Reinforcement Learning, MMSR treats the mapping from data to expressions as a translation problem and introduces large-scale pre-trained models for modal alignment. The approach leverages contrastive learning to facilitate modal feature fusion, training multiple losses simultaneously for better feature extraction and fusion. Experimental results demonstrate MMSR’s superiority over baselines on mainstream datasets, including SRBench. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary MMSR is a new way to solve the problem of symbolic regression. This means finding simple formulas that describe complex laws in nature. For a long time, scientists have tried to use machines to find these formulas, but it was hard because they had to teach the machines what the formulas should look like. MMSR makes it easier by treating this as a language translation problem, where the machine learns to translate data into formulas. This is achieved through large-scale pre-training and contrastive learning. The result is that MMSR can find better formulas than other methods on popular datasets. |
Keywords
* Artificial intelligence * Alignment * Feature extraction * Multi modal * Regression * Reinforcement learning * Translation
