Directed Graph Grammars for Sequence-based Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sun, Michael, Foo, Orion, Liu, Gang, Matusik, Wojciech, Chen, Jie
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912400924475392
author Sun, Michael
Foo, Orion
Liu, Gang
Matusik, Wojciech
Chen, Jie
author_facet Sun, Michael
Foo, Orion
Liu, Gang
Matusik, Wojciech
Chen, Jie
contents Directed acyclic graphs (DAGs) are a class of graphs commonly used in practice, with examples that include electronic circuits, Bayesian networks, and neural architectures. While many effective encoders exist for DAGs, it remains challenging to decode them in a principled manner, because the nodes of a DAG can have many different topological orders. In this work, we propose a grammar-based approach to constructing a principled, compact and equivalent sequential representation of a DAG. Specifically, we view a graph as derivations over an unambiguous grammar, where the DAG corresponds to a unique sequence of production rules. Equivalently, the procedure to construct such a description can be viewed as a lossless compression of the data. Such a representation has many uses, including building a generative model for graph generation, learning a latent space for property prediction, and leveraging the sequence representational continuity for Bayesian Optimization over structured data. Code is available at https://github.com/shiningsunnyday/induction.
format Preprint
id arxiv_https___arxiv_org_abs_2505_22949
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Directed Graph Grammars for Sequence-based Learning
Sun, Michael
Foo, Orion
Liu, Gang
Matusik, Wojciech
Chen, Jie
Machine Learning
Directed acyclic graphs (DAGs) are a class of graphs commonly used in practice, with examples that include electronic circuits, Bayesian networks, and neural architectures. While many effective encoders exist for DAGs, it remains challenging to decode them in a principled manner, because the nodes of a DAG can have many different topological orders. In this work, we propose a grammar-based approach to constructing a principled, compact and equivalent sequential representation of a DAG. Specifically, we view a graph as derivations over an unambiguous grammar, where the DAG corresponds to a unique sequence of production rules. Equivalently, the procedure to construct such a description can be viewed as a lossless compression of the data. Such a representation has many uses, including building a generative model for graph generation, learning a latent space for property prediction, and leveraging the sequence representational continuity for Bayesian Optimization over structured data. Code is available at https://github.com/shiningsunnyday/induction.
title Directed Graph Grammars for Sequence-based Learning
topic Machine Learning
url https://arxiv.org/abs/2505.22949