# Accelerating Genetic Programming using GPUs

arXiv: https://arxiv.org/abs/2110.11226 · HTML: https://mlsanity.com/paper/2110.11226/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 24 references that are in the map.
- First posted on arXiv 15 Oct 2021.

## Builds on (most cited first, 4 of 4)

- [Machine Learning in Python: Main developments and technology trends in data science, machine learning, and artificial intelligence](https://mlsanity.com/paper/2002.04803/) (2020)
- [TensorFlow Enabled Genetic Programming](https://mlsanity.com/paper/1708.03157/) (2017)
- [TensorGP -- Genetic Programming Engine in TensorFlow](https://mlsanity.com/paper/2103.07512/) (2021)
- [Speed Benchmarking of Genetic Programming Frameworks](https://mlsanity.com/paper/2106.11919/) (2021)

## Built on by (most cited first, 3 of 3)

- [A computational framework for physics-informed symbolic regression with straightforward integration of domain knowledge](https://mlsanity.com/paper/2209.06257/) (2022)
- [Symbolic Regression as Feature Engineering Method for Machine and Deep Learning Regression Tasks](https://mlsanity.com/paper/2311.06028/) (2023)
- [Kozax: Flexible and Scalable Genetic Programming in JAX](https://mlsanity.com/paper/2502.03047/) (2025)

## Abstract (arXiv)

Genetic Programming (GP), an evolutionary learning technique, has multiple applications in machine learning such as curve fitting, data modelling, feature selection, classification etc. GP has several inherent parallel steps, making it an ideal candidate for GPU based parallelization. This paper describes a GPU accelerated stack-based variant of the generational GP algorithm which can be used for symbolic regression and binary classification. The selection and evaluation steps of the generational GP algorithm are parallelized using CUDA. We introduce representing candidate solution expressions as prefix lists, which enables evaluation using a fixed-length stack in GPU memory. CUDA based matrix vector operations are also used for computation of the fitness of population programs. We evaluate our algorithm on synthetic datasets for the Pagie Polynomial (ranging in size from 4096 to 16 million points), profiling training times of our algorithm with other standard symbolic regression libraries viz. gplearn, TensorGP and KarooGP. In addition, using 6 large-scale regression and classification datasets usually used for comparing gradient boosting algorithms, we run performance benchmarks on our algorithm and gplearn, profiling the training time, test accuracy, and loss. On an NVIDIA DGX-A100 GPU, our algorithm outperforms all the previously listed frameworks, and in particular, achieves average speedups of 119× and 40× against gplearn on the synthetic and large scale datasets respectively.