FicGCN: Unveiling the Homomorphic Encryption Efficiency from Irregular Graph Convolutional Networks

2025-06-12

AI Summary

To address the high computational overhead and poor exploitability of sparsity in homomorphic encryption (HE)-based graph convolutional networks (GCNs), this paper proposes an efficient HE-GCN acceleration framework for privacy-preserving graph learning. The method introduces two key innovations: (1) Sparse ciphertext intra-aggregation (SpIntra-CA), which directly aggregates neighbor information within the ciphertext packing space, eliminating redundant rotation operations; and (2) neighborhood-driven region reordering, jointly optimizing adjacency-aware data rearrangement and low-latency rotations to mitigate irregular memory access costs. The framework tightly integrates HE principles, sparse graph representations, and ciphertext-level computation optimizations. Evaluated on multiple benchmark datasets, the proposed approach achieves up to 4.10× end-to-end inference speedup over state-of-the-art HE-GCN methods, significantly advancing the practicality of encrypted graph neural networks.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Heuristic Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Security and Privacy: Applications of cryptography
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Graph Convolutional Neural Networks (GCNs) have gained widespread popularity in various fields like personal healthcare and financial systems, due to their remarkable performance. Despite the growing demand for cloud-based GCN services, privacy concerns over sensitive graph data remain significant. Homomorphic Encryption (HE) facilitates Privacy-Preserving Machine Learning (PPML) by allowing computations to be performed on encrypted data. However, HE introduces substantial computational overhead, particularly for GCN operations that require rotations and multiplications in matrix products. The sparsity of GCNs offers significant performance potential, but their irregularity introduces additional operations that reduce practical gains. In this paper, we propose FicGCN, a HE-based framework specifically designed to harness the sparse characteristics of GCNs and strike a globally optimal balance between aggregation and combination operations. FicGCN employs a latency-aware packing scheme, a Sparse Intra-Ciphertext Aggregation (SpIntra-CA) method to minimize rotation overhead, and a region-based data reordering driven by local adjacency structure. We evaluated FicGCN on several popular datasets, and the results show that FicGCN achieved the best performance across all tested datasets, with up to a 4.10x improvement over the latest design.

Problem

• Reducing computational overhead in Homomorphic Encryption for GCNs
• Optimizing sparse GCN operations to enhance privacy-preserving performance
• Balancing aggregation and combination operations for efficient HE-based GCNs

Innovation

• Latency-aware packing scheme for efficiency
• Sparse Intra-Ciphertext Aggregation reduces rotation overhead
• Region-based data reordering optimizes adjacency structure