Saved in:
Bibliographic Details
Main Authors: Piquenot, Jason, Bérar, Maxime, Héroux, Pierre, Ramel, Jean-Yves, Raveaux, Romain, Adam, Sébastien
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.01661
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916578773172224
author Piquenot, Jason
Bérar, Maxime
Héroux, Pierre
Ramel, Jean-Yves
Raveaux, Romain
Adam, Sébastien
author_facet Piquenot, Jason
Bérar, Maxime
Héroux, Pierre
Ramel, Jean-Yves
Raveaux, Romain
Adam, Sébastien
contents This paper presents Grammar Reinforcement Learning (GRL), a reinforcement learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer architecture that models a Pushdown Automaton (PDA) within a context-free grammar (CFG) framework. Taking as use case the problem of efficiently counting paths and cycles in graphs, a key challenge in network analysis, computer science, biology, and social sciences, GRL discovers new matrix-based formulas for path/cycle counting that improve computational efficiency by factors of two to six w.r.t state-of-the-art approaches. Our contributions include: (i) a framework for generating gramformers that operate within a CFG, (ii) the development of GRL for optimizing formulas within grammatical structures, and (iii) the discovery of novel formulas for graph substructure counting, leading to significant computational improvements.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01661
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Finding path and cycle counting formulae in graphs with Deep Reinforcement Learning
Piquenot, Jason
Bérar, Maxime
Héroux, Pierre
Ramel, Jean-Yves
Raveaux, Romain
Adam, Sébastien
Artificial Intelligence
Formal Languages and Automata Theory
This paper presents Grammar Reinforcement Learning (GRL), a reinforcement learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer architecture that models a Pushdown Automaton (PDA) within a context-free grammar (CFG) framework. Taking as use case the problem of efficiently counting paths and cycles in graphs, a key challenge in network analysis, computer science, biology, and social sciences, GRL discovers new matrix-based formulas for path/cycle counting that improve computational efficiency by factors of two to six w.r.t state-of-the-art approaches. Our contributions include: (i) a framework for generating gramformers that operate within a CFG, (ii) the development of GRL for optimizing formulas within grammatical structures, and (iii) the discovery of novel formulas for graph substructure counting, leading to significant computational improvements.
title Finding path and cycle counting formulae in graphs with Deep Reinforcement Learning
topic Artificial Intelligence
Formal Languages and Automata Theory
url https://arxiv.org/abs/2410.01661