Complexity-Aware Deep Symbolic Regression with Robust Risk-Seeking Policy Gradients

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bastiani, Zachary, Kirby, Robert M., Hochhalter, Jacob, Zhe, Shandian
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917363308298240
author Bastiani, Zachary
Kirby, Robert M.
Hochhalter, Jacob
Zhe, Shandian
author_facet Bastiani, Zachary
Kirby, Robert M.
Hochhalter, Jacob
Zhe, Shandian
contents We propose a novel deep symbolic regression approach to enhance the robustness and interpretability of data-driven mathematical expression discovery. Our work is aligned with the popular DSR framework which focuses on learning a data-specific expression generator, without relying on pretrained models or additional search or planning procedures. Despite the success of existing DSR methods, they are built on recurrent neural networks, solely guided by data fitness, and potentially meet tail barriers that can zero out the policy gradient, causing inefficient model updates. To overcome these limitations, we design a decoder-only architecture that performs attention in the frequency domain and introduce a dual-indexed position encoding to conduct layer-wise generation. Second, we propose a Bayesian information criterion (BIC)-based reward function that can automatically adjust the trade-off between expression complexity and data fitness, without the need for explicit manual tuning. Third, we develop a ranking-based weighted policy update method that eliminates the tail barriers and enhances training effectiveness. Extensive benchmarks and systematic experiments demonstrate the advantages of our approach. We have released our implementation at https://github.com/ZakBastiani/CADSR.
format Preprint
id arxiv_https___arxiv_org_abs_2406_06751
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Complexity-Aware Deep Symbolic Regression with Robust Risk-Seeking Policy Gradients
Bastiani, Zachary
Kirby, Robert M.
Hochhalter, Jacob
Zhe, Shandian
Machine Learning
Artificial Intelligence
We propose a novel deep symbolic regression approach to enhance the robustness and interpretability of data-driven mathematical expression discovery. Our work is aligned with the popular DSR framework which focuses on learning a data-specific expression generator, without relying on pretrained models or additional search or planning procedures. Despite the success of existing DSR methods, they are built on recurrent neural networks, solely guided by data fitness, and potentially meet tail barriers that can zero out the policy gradient, causing inefficient model updates. To overcome these limitations, we design a decoder-only architecture that performs attention in the frequency domain and introduce a dual-indexed position encoding to conduct layer-wise generation. Second, we propose a Bayesian information criterion (BIC)-based reward function that can automatically adjust the trade-off between expression complexity and data fitness, without the need for explicit manual tuning. Third, we develop a ranking-based weighted policy update method that eliminates the tail barriers and enhances training effectiveness. Extensive benchmarks and systematic experiments demonstrate the advantages of our approach. We have released our implementation at https://github.com/ZakBastiani/CADSR.
title Complexity-Aware Deep Symbolic Regression with Robust Risk-Seeking Policy Gradients
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2406.06751