Asynchronous Event Error-Minimizing Noise for Safeguarding Event Dataset

2025-07-08

AI Summary

To address the vulnerability of event camera data to unauthorized model training, this paper proposes the first unlearnable sample generation method tailored for asynchronous event streams. Our approach introduces a sparsity-aware noise projection strategy that injects adversarial perturbations via an asynchronous event error minimization mechanism—preserving the intrinsic sparsity of event data while steering illicit models to converge on noise patterns rather than genuine semantics. This work pioneers the extension of the unlearnable sample paradigm to spike-based event stream learning, uniquely balancing privacy protection and downstream task utility. Extensive experiments across multiple event-based recognition benchmarks demonstrate that our method reduces unauthorized training accuracy by over 80%, while degrading legitimate user performance by less than 2%.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Security and Privacy: Data transparency and provenance
• 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

With more event datasets being released online, safeguarding the event dataset against unauthorized usage has become a serious concern for data owners. Unlearnable Examples are proposed to prevent the unauthorized exploitation of image datasets. However, it's unclear how to create unlearnable asynchronous event streams to prevent event misuse. In this work, we propose the first unlearnable event stream generation method to prevent unauthorized training from event datasets. A new form of asynchronous event error-minimizing noise is proposed to perturb event streams, tricking the unauthorized model into learning embedded noise instead of realistic features. To be compatible with the sparse event, a projection strategy is presented to sparsify the noise to render our unlearnable event streams (UEvs). Extensive experiments demonstrate that our method effectively protects event data from unauthorized exploitation, while preserving their utility for legitimate use. We hope our UEvs contribute to the advancement of secure and trustworthy event dataset sharing. Code is available at: https://github.com/rfww/uevs.

Problem

• Prevent unauthorized training from event datasets
• Generate unlearnable asynchronous event streams
• Protect event data while preserving legitimate utility

Innovation

• Asynchronous event error-minimizing noise generation
• Sparse noise projection for event streams
• Unlearnable event streams prevent unauthorized training