# Active Learning Improves Performance on Symbolic RegressionTasks in StackGP

arXiv: https://arxiv.org/abs/2202.04708 · HTML: https://mlsanity.com/paper/2202.04708/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 16 references that are in the map.
- First posted on arXiv 9 Feb 2022.

## Builds on (most cited first, 2 of 2)

- [AI Feynman: a Physics-Inspired Method for Symbolic Regression](https://mlsanity.com/paper/1905.11481/) (2019)
- [Fitness Landscape Analysis of Dimensionally-Aware Genetic Programming Featuring Feynman Equations](https://mlsanity.com/paper/2004.12762/) (2020)

## Built on by (most cited first, 8 of 8)

- [Neuro-Symbolic AI in 2024: A Systematic Review](https://mlsanity.com/paper/2501.05435/) (2025)
- [LLM-ACES: Closed-Loop Discovery of Dynamical Systems with LLM-Guided Adaptive Search](https://mlsanity.com/paper/2606.25039/) (2026)
- [Vertical Symbolic Regression via Deep Policy Gradient](https://mlsanity.com/paper/2402.00254/) (2024)
- [Active Learning in Symbolic Regression with Physical Constraints](https://mlsanity.com/paper/2305.10379/) (2023)
- [Active Learning in Genetic Programming: Guiding Efficient Data Collection for Symbolic Regression](https://mlsanity.com/paper/2308.00672/) (2023)
- [Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching](https://mlsanity.com/paper/2409.01416/) (2024)
- [Online Symbolic Regression with Informative Query](https://mlsanity.com/paper/2302.10539/) (2023)
- [Vertical Symbolic Regression](https://mlsanity.com/paper/2312.11955/) (2023)

## Abstract (arXiv)

In this paper we introduce an active learning method for symbolic regression using StackGP. The approach begins with a small number of data points for StackGP to model. To improve the model the system incrementally adds a data point such that the new point maximizes prediction uncertainty as measured by the model ensemble. Symbolic regression is re-run with the larger data set. This cycle continues until the system satisfies a termination criterion. We use the Feynman AI benchmark set of equations to examine the ability of our method to find appropriate models using fewer data points. The approach was found to successfully rediscover 72 of the 100 Feynman equations using as few data points as possible, and without use of domain expertise or data translation.