Heterogeneous Graph Neural Architecture Search with GPT-4

TL;DR

This paper introduces GHGNAS, a GPT-4 based method for heterogeneous graph neural architecture search that improves efficiency, stability, and accuracy over previous algorithms by leveraging prompt-based iterative optimization.

Contribution

The paper proposes a novel GPT-4 guided HGNAS approach that enhances search stability and effectiveness through prompt-based feedback loops.

Findings

GHGNAS outperforms previous HGNAS methods in efficiency and stability.

The GPT-4 based approach effectively designs new heterogeneous graph neural networks.

Experimental results demonstrate improved search accuracy and robustness.

Abstract

Heterogeneous graph neural architecture search (HGNAS) represents a powerful tool for automatically designing effective heterogeneous graph neural networks. However, existing HGNAS algorithms suffer from inefficient searches and unstable results. In this paper, we present a new GPT-4 based HGNAS model to improve the search efficiency and search accuracy of HGNAS. Specifically, we present a new GPT-4 enhanced Heterogeneous Graph Neural Architecture Search (GHGNAS for short). The basic idea of GHGNAS is to design a set of prompts that can guide GPT-4 toward the task of generating new heterogeneous graph neural architectures. By iteratively asking GPT-4 with the prompts, GHGNAS continually validates the accuracy of the generated HGNNs and uses the feedback to further optimize the prompts. Experimental results show that GHGNAS can design new HGNNs by leveraging the powerful generalization capability of GPT-4. Moreover, GHGNAS runs more effectively and stably than previous HGNAS models based on reinforcement learning and differentiable search algorithms.