Contrastive Learning for Continuous Touch-Based Authentication

2025-04-24

AI Summary

To address the trade-off between naturalness and security in continuous touch-based authentication on mobile devices, this paper proposes the first end-to-end fine-grained authentication framework based on contrastive learning. Methodologically, it innovatively integrates a Temporal Masked Autoencoder (TMAE) with a dual-branch temporal attention convolutional network to jointly model touch dynamics and user-specific behavioral discriminability; it further introduces, for the first time, a synergistic mechanism combining multi-head attention and channel-wise attention to enhance spatiotemporal feature representation. Evaluated on both public benchmarks and a newly constructed dataset, the method achieves state-of-the-art performance: +5.2% authentication accuracy, −38% false rejection rate, and millisecond-level real-time inference. This work establishes a new paradigm for seamless, continuous, and highly robust behavioral biometric authentication on mobile platforms.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Life-Long and Continual Learning
• Humans and AI: Interaction Techniques and Devices

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Smart mobile devices have become indispensable in modern daily life, where sensitive information is frequently processed, stored, and transmitted-posing critical demands for robust security controls. Given that touchscreens are the primary medium for human-device interaction, continuous user authentication based on touch behavior presents a natural and seamless security solution. While existing methods predominantly adopt binary classification under single-modal learning settings, we propose a unified contrastive learning framework for continuous authentication in a non-disruptive manner. Specifically, the proposed method leverages a Temporal Masked Autoencoder to extract temporal patterns from raw multi-sensor data streams, capturing continuous motion and gesture dynamics. The pre-trained TMAE is subsequently integrated into a Siamese Temporal-Attentive Convolutional Network within a contrastive learning paradigm to model both sequential and cross-modal patterns. To further enhance performance, we incorporate multi-head attention and channel attention mechanisms to capture long-range dependencies and optimize inter-channel feature integration. Extensive experiments on public benchmarks and a self-collected dataset demonstrate that our approach outperforms state-of-the-art methods, offering a reliable and effective solution for user authentication on mobile devices.

Problem

• Develop continuous touch-based authentication for mobile security
• Overcome limitations of single-modal binary classification methods
• Enhance authentication via multi-sensor temporal and cross-modal patterns

Innovation

• Contrastive learning framework for continuous authentication
• Temporal Masked Autoencoder extracts touch dynamics
• Siamese network with attention mechanisms enhances performance