Approximate Subgraph Matching with Neural Graph Representations and Reinforcement Learning

IEEE Publication Technology, Kaiyang Li, Shihao Ji, Zhipeng Cai, Wei Li · 2026

Read Approximate Subgraph Matching with Neural Graph Representations and Reinforcement Learning by IEEE Publication Technology, Kaiyang Li, Shihao Ji, Zhipeng Cai, Wei Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18314.txt.

Description

Li and co-authors propose RL-ASM, a reinforcement learning method for approximate subgraph matching. The task checks whether a query graph appears, roughly, inside a large target graph. Heuristic searches miss much graph information and settle for weak matches. A graph transformer encodes both graphs fully to guide branch-and-bound node pairing. Imitation learning warms up the policy, then PPO fine-tuning lifts long-term reward; tests beat past methods on speed and quality.

Catalog record

CREATORIEEE Publication Technology, Kaiyang Li, Shihao Ji, Zhipeng Cai, Wei Li
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE51810
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.18314

Browse the full library · About the Tower of Babel