Scaling Up Unbiased Search-based Symbolic Regression
Authors: Paul Kahlmeyer, Joachim Giesen, Michael Habeck, Henrik Voigt
IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conducted two types of experiments. In the first experiment, we compare the performance of our unbiased, searchbased approach to the state of the art in symbolic regression on the established and comprehensive SRBench test suite by [La Cava et al., 2021]. In the second experiment, we evaluate the scalability advantage that variable augmentation brings to the unbiased, search-based approach. |
| Researcher Affiliation | Academia | 1Friedrich Schiller University Jena 2University Hospital Jena {paul.kahlmeyer, joachim.giesen, michael.habeck, henrik.voigt}@uni-jena.de |
| Pseudocode | No | The paper describes algorithms in text and provides figures illustrating concepts (like Figure 4 for the algorithm flow), but it does not include a formal pseudocode block or algorithm listing. |
| Open Source Code | No | The paper does not provide an explicit statement or link for open-sourcing the code for their method. |
| Open Datasets | Yes | We conducted two types of experiments. In the first experiment, we compare the performance of our unbiased, searchbased approach to the state of the art in symbolic regression on the established and comprehensive SRBench test suite by [La Cava et al., 2021]. |
| Dataset Splits | No | The paper refers to using training data, but does not specify explicit train/validation/test splits with percentages or counts, or refer to standard splits for the SRBench test suite beyond general evaluation on test data. |
| Hardware Specification | Yes | All experiments were run on a computer with an Intel Xeon Gold 6226R 64-core processor, 128 GB of RAM, and Python 3.10. |
| Software Dependencies | Yes | All experiments were run on a computer with an Intel Xeon Gold 6226R 64-core processor, 128 GB of RAM, and Python 3.10. The symbolical checks are delegated to the Python library Sym Py [Meurer et al., 2017]. |
| Experiment Setup | Yes | Unless stated otherwise, our method, named UDFS (Unbiased DAG Frame Search) was used with five intermediary nodes and a maximum of 200 000 DAG skeletons. For the variable augmentation, we have used polynomial regression (UDFS + Aug) to select k = 1 augmentations and up to 30 nodes in the corresponding DAG search. |