Inductive meta-path learning for schema-complex heterogeneous information networks (SchemaWalk)

This repository contains a PyTorch implementation of SchemaWalk, as described in the paper:

Shixuan Liu, Changjun Fan, Kewei Cheng, Yunfei Wang, Peng Cui, Yizhou Sun, Zhong Liu. Inductive meta-path learning for schema-complex heterogeneous information networks. IEEE Transactions on Pattern Analysis and Machine Intelligence (2024).

Overview

Heterogeneous Information Networks (HINs) are information networks with multiple types of nodes and edges. The concept of meta-path, i.e., a sequence of entity types and relation types connecting two entities, is proposed to provide the meta-level explainable semantics for various HIN tasks. Traditionally, meta-paths are primarily used for schema-simple HINs, e.g., bibliographic networks with only a few entity types, where meta-paths are often enumerated with domain knowledge. However, the adoption of meta-paths for schema-complex HINs, such as knowledge bases (KBs) with hundreds of entity and relation types, has been limited due to the computational complexity associated with meta-path enumeration. Additionally, effectively assessing meta-paths requires enumerating relevant path instances, which adds further complexity to the meta-path learning process.To address these challenges, we propose~\model, an inductive meta-path learning framework for schema-complex HINs. We represent meta-paths with schema-level representations to support the learning of the scores of meta-paths for varying relations, mitigating the need of exhaustive path instance enumeration for each relation. Further, we design a reinforcement-learning based path-finding agent, which directly navigates the network schema (i.e., schema graph) to learn policies for establishing meta-paths with high coverage and confidence for multiple relations. Extensive experiments on real data sets demonstrate the effectiveness of our proposed paradigm.

System Requirements

Recommended software versions:

python==3.8
pytorch==1.12.1
numpy==1.23.4
scipy==1.3.1

Training Instructions

Multi-Relation Inductive Setting

Example for Yago/DBpedia:

python train.py --base_iterations 500 --beta 0.1 --hc_coff 10.0 --hidden_size 200 --reward_device auto --global_coverage --gpu 0
python train.py --base_iterations 500 --beta 0.05 --hc_coff 5.0 --hidden_size 400 --reward_device auto --global_coverage --gpu 0

Multi-Relation Transductive Setting

Example for Yago/DBpedia:

python train.py --base_iterations 500 --beta 0.1 --hc_coff 10.0 --hidden_size 200 --reward_device auto --global_coverage --gpu 0 --transductive
python train.py --base_iterations 500 --beta 0.05 --hc_coff 5.0 --hidden_size 400 --reward_device auto --global_coverage --gpu 0 --transductive

Per-Relation Inductive Setting

Example:

python train.py --gpu 0 --base_iterations 1000 --reward_device cpu --target_rel_index 1 --hc_coff 10 --beta 0.05

Adjust target_rel_index according to the index of the target relation in the relation set (see train.py).

Trained Models and Meta-Paths

Some pre-trained models and outputted meta-paths are available in ./output/ and ./metapaths/.

Reference

If you find our code or paper useful, please cite:

@article{liu2024inductive,
  title={Inductive meta-path learning for schema-complex heterogeneous information networks},
  author={Liu, Shixuan and Fan, Changjun and Cheng, Kewei and Wang, Yunfei and Cui, Peng and Sun, Yizhou and Liu, Zhong},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2024},
  publisher={IEEE}
}