DESIGN: Encrypted GNN Inference via Server-Side Input Graph Pruning

TL;DR

DESIGN introduces a server-side hierarchical optimization framework that significantly accelerates encrypted GNN inference by pruning input graphs and adapting activation complexity, enabling practical privacy-preserving graph analytics.

Contribution

It presents a novel FHE-compatible input graph pruning and adaptive activation scheme, improving efficiency of encrypted GNN inference over existing methods.

Findings

Substantially faster FHE GNN inference compared to state-of-the-art.

Maintains competitive accuracy while optimizing performance.

Effective input graph pruning guided by encrypted importance scores.

Abstract

Graph Neural Networks (GNNs) have achieved state-of-the-art performance in various graph-based learning tasks. However, enabling privacy-preserving GNNs in encrypted domains, such as under Fully Homomorphic Encryption (FHE), typically incurs substantial computational overhead, rendering real-time and privacy-preserving inference impractical. In this work, we propose DESIGN (EncrypteD GNN Inference via sErver-Side Input Graph pruNing), a novel framework for efficient encrypted GNN inference. DESIGN tackles the critical efficiency limitations of existing FHE GNN approaches, which often overlook input data redundancy and apply uniform computational strategies. Our framework achieves significant performance gains through a hierarchical optimization strategy executed entirely on the server: first, FHE-compatible node importance scores (based on encrypted degree statistics) are computed from the encrypted graph. These scores then guide a homomorphic partitioning process, generating multi-level importance masks directly under FHE. This dynamically generated mask facilitates both input graph pruning (by logically removing unimportant elements) and a novel adaptive polynomial activation scheme, where activation complexity is tailored to node importance levels. Empirical evaluations demonstrate that DESIGN substantially accelerates FHE GNN inference compared to state-of-the-art methods while maintaining competitive model accuracy, presenting a robust solution for secure graph analytics. Our implementation is publicly available at https://github.com/LabRAI/DESIGN.