Vertical Symbolic Regression via Deep Policy Gradient

Authors: Nan Jiang, Md Nasim, Yexiang Xue

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate that our VSR-DPG significantly outperforms popular baselines in identifying both algebraic equations and ordinary differential equations on a series of benchmarks. (1) In Table 1, our VSR-DPG attains the smallest median NMSE values in 7 out of 8 datasets, against current popular baselines including VSR-GP. (2) Further analysis on the best-discovered equation (in Table 3) shows that VSR-DPG uncovers up to 50% of the exact governing equations with 5 input variables, where the baselines only attain 0%. (3) In Table 2, our VSR-DPG can find high-quality expressions on datasets with up to 50 variables, because of the vertical discovery idea. (4) On discovery of ordinary differential equations in Table 4, our VSR-DPG also improves over current baselines.1
Researcher Affiliation Academia Nan Jiang , Md Nasim , Yexiang Xue Department of Computer Science, Purdue University {jiang631, mnasim, yexiang}@purdue.edu
Pseudocode Yes We summarize the whole process of VSR-DPG in Algorithm 1 in the appendix.
Open Source Code Yes 1The code is at https://github.com/jiangnanhugo/VSR-DPG.
Open Datasets Yes For the dataset on algebraic expressions, we consider the 8 groups of expressions from the Trigonometric dataset [Jiang and Xue, 2023]... All of them are collected from [Brunton et al., 2016].
Dataset Splits No The paper mentions evaluating on a 'separately-generated testing dataset' but does not provide specific train/validation/test splits (e.g., percentages or sample counts) or references to predefined splits for reproduction.
Hardware Specification No The paper does not provide any specific details regarding the hardware used for running the experiments (e.g., CPU/GPU models, memory, or specific computing infrastructure).
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., 'Python 3.8', 'PyTorch 1.9').
Experiment Setup Yes The detailed settings are in Appendix C.2. ... The detailed experiment configurations are in Appendix C.4.