Neural-Evolutionary Symbolic Regression with Global Constraints: Constraint-Aware Decoding and Reward Shaping

Created by W.Langdon from gp-bibliography.bib Revision:1.9085

@InProceedings{Xiangdong_Wu:2026:ICML,

• author = "Xiangdong Wu and Wenjun Wu and Ziyu Wei and Bingrun Chen and Zhenbo Song and Rongye Shi",
• title = "Neural-Evolutionary Symbolic Regression with Global Constraints: Constraint-Aware Decoding and Reward Shaping",
• booktitle = "Proceedings of the 43rd International Conference on Machine Learning",
• year = "2026",
• editor = "Alekh Agarwal and Miro Dudik and Martin Jaggi and Sharon Li",
• pages = "PMLR 306",
• address = "Seoul, South Korea",
• month = jul # " 6-11",
• keywords = "genetic algorithms, genetic programming, ANN, Symbolic Perfect Binary Trees (SPBTs)",
• URL = "https://icml.cc/virtual/2026/poster/62673",
• abstract = "Symbolic regression discovers interpretable mathematical expressions from data and is central to scientific modeling. Recent neural approaches typically linearise expression trees into token sequences for sequential generation, but this representation weakens access to the underlying hierarchy and makes it difficult to enforce structure-dependent constraints. Hybrid neural--evolutionary frameworks further combine neural generators with genetic programming (GP), yet training can be unstable due to distribution mismatch between neural samples and GP-refined elites. We propose GCN-SR, a graph-based symbolic regression framework that generates expressions directly in an explicit tree form. GCN-SR introduces Symbolic Perfect Binary Trees (SPBTs), a fixed-topology scaffold that enables batched tree generation and supports an autoregressive generator based on a Graph Convolutional Network (GCN) while preserving hierarchical structure. To leverage GP refinement without unstable target matching, we further introduce Similarity-Weighted Policy Gradient (SWPG), which uses GP only to construct similarity-weighted reward signals. Experiments on standard symbolic regression benchmarks, together with extensive ablations, show that GCN-SR consistently outperforms strong neural and hybrid baselines.",
• notes = "ICML 2026 https://icml.cc/virtual/2026",

}

Genetic Programming entries for Xiangdong Wu Wenjun Wu Ziyu Wei Bingrun Chen Zhenbo Song Rongye Shi

Citations