Articles
SN Computer Science (2662995X)6(6)
The rise of cloud computing has transformed how we process and analyse data, particularly in the domain of machine learning as a service (MLaaS). Protecting data privacy and proprietary models has become paramount in this evolving landscape. The challenge lies in ensuring accurate and reliable inference while safeguarding sensitive elements such as model parameters (weights and biases) and client data. The security landscape has traditionally relied on cryptographic approaches, including garbled circuits (GC), homomorphic encryption (HE), and oblivious transfer (OT), to protect inference processes. However, the emergence of function secret sharing (FSS) has introduced a more streamlined approach, offering reduced computational and communicatio n complexity. While FSS has proven effective for secure inference under semi-honest threat models, it faces a significant limitation: its dependence on the assumption that the trusted third party (TTP) will not engage in collusion with other participants. This assumption represents a potential vulnerability in the system’s security framework. We thoroughly examine various secure inference schemes for neural networks (NNs). By examining and comparing the strengths and limitations of each scheme, we aim to provide researchers with valuable insights into artificial intelligence security. This comparative analysis is a resource for those working in related fields, particularly in neural networks, helping them make informed decisions about security implementations in their research and applications. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.
Peer-to-Peer Networking and Applications (19366450)18(4)
Mobile devices have become essential to our daily lives, leading to a growing need for robust security mechanisms in their communications. Ensuring secure interactions between these devices and central servers is vital to protect sensitive data. As a result, there is a significant demand for Authenticated Key Exchange (AKE) schemes. Schemes that rely on passwords for authentication and key exchange are known as Password Authenticated Key Exchange (PAKE). The development of Shor’s algorithm in 1994, along with recent advances in quantum computing, has led to researchers to propose schemes, including PAKE, that are secure against quantum attacks. Recently, Moony et al. introduced a lattice-based two-party authentication protocol for mobile devices. In this paper, we analyze the vulnerabilities of their scheme, focusing on key mismatch attack, forward secrecy violation, replay attack, Key Compromise Impersonation (KCI) attack, and offline password guessing attack. To address these issues, we propose a new reconciliation-based anonymous PAKE scheme based on RLWE, secure in the random oracle model. Our scheme is not only resistant to signal leakage and key mismatch attacks, which affect existing reconciliation-based RLWE key exchange protocols, but also uniquely ensures KCI resistance−a property that is not provided by prior anonymous PAKE schemes. The results show that while our scheme provides the strongest security, user-side computational complexity is reduced by about 5% compared to the most secure scheme, based on benchmark parameters suited for mobile environments. Additionally, it incurs approximately 11% higher communication overhead compared to most schemes. Despite these trade-offs, the significant security improvements make our scheme highly suitable for mobile applications, where user-side efficiency is critical. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.
