Efficient Private Inference Based on Helper-Assisted Malicious Security Dishonest Majority MPC

2025-07-13

AI Summary

To address the low efficiency and weak security of maliciously secure multi-party computation (MPC) for private inference in machine learning as a service (MLaaS) under a dishonest-majority threat model, this paper proposes the first assistant-aided maliciously secure MPC framework supporting a dishonest majority. Our approach introduces five fixed-round protocols and a co-optimization strategy, integrating sixth-order polynomial approximation for nonlinear activation functions and tunable-parameter batch normalization to effectively suppress activation value leakage while preserving high-fidelity function fitting. Evaluated on LeNet and AlexNet, our framework achieves 2.4–25.7× speedup over baselines under LAN and 1.3–9.5× under WAN, with relative errors as low as 0.04%–1.08%. To the best of our knowledge, this is the first work to simultaneously achieve high efficiency and high accuracy in neural network inference under strong malicious-security guarantees.

Technology Category

• Machine Learning: Hardware-aware ML
• Natural Language Processing: Safety and Robustness
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Security and Privacy: Security and privacy of machine learning and AI applications
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Private inference based on Secure Multi-Party Computation (MPC) addresses data privacy risks in Machine Learning as a Service (MLaaS). However, existing MPC-based private inference frameworks focuses on semi-honest or honest majority models, whose threat models are overly idealistic, while malicious security dishonest majority models face the challenge of low efficiency. To balance security and efficiency, we propose a private inference framework using Helper-Assisted Malicious Security Dishonest Majority Model (HA-MSDM). This framework includes our designed five MPC protocols and a co-optimized strategy. These protocols achieve efficient fixed-round multiplication, exponentiation, and polynomial operations, providing foundational primitives for private inference. The co-optimized strategy balances inference efficiency and accuracy. To enhance efficiency, we employ polynomial approximation for nonlinear layers. For improved accuracy, we construct sixth-order polynomial approximation within a fixed interval to achieve high-precision activation function fitting and introduce parameter-adjusted batch normalization layers to constrain the activation escape problem. Benchmark results on LeNet and AlexNet show our framework achieves 2.4-25.7x speedup in LAN and 1.3-9.5x acceleration in WAN compared to state-of-the-art frameworks (IEEE S&P'25), maintaining high accuracy with only 0.04%-1.08% relative errors.

Problem

• Balancing security and efficiency in private inference frameworks
• Enhancing MPC protocols for malicious security dishonest majority models
• Improving accuracy and speed in MLaaS private inference

Innovation

• Helper-Assisted Malicious Security Dishonest Majority Model
• Five efficient fixed-round MPC protocols
• Polynomial approximation for nonlinear layers