PriMask: Cascadable and Collusion-Resilient Data Masking for Mobile Cloud Inference
arXiv:2211.06716 · doi:10.1145/3560905.3568531
Abstract
Mobile cloud offloading is indispensable for inference tasks based on large-scale deep models. However, transmitting privacy-rich inference data to the cloud incurs concerns. This paper presents the design of a system called PriMask, in which the mobile device uses a secret small-scale neural network called MaskNet to mask the data before transmission. PriMask significantly weakens the cloud's capability to recover the data or extract certain private attributes. The MaskNet is em cascadable in that the mobile can opt in to or out of its use seamlessly without any modifications to the cloud's inference service. Moreover, the mobiles use different MaskNets, such that the collusion between the cloud and some mobiles does not weaken the protection for other mobiles. We devise a {\em split adversarial learning} method to train a neural network that generates a new MaskNet quickly (within two seconds) at run time. We apply PriMask to three mobile sensing applications with diverse modalities and complexities, i.e., human activity recognition, urban environment crowdsensing, and driver behavior recognition. Results show PriMask's effectiveness in all three applications.
arXiv comment13 pages, best paper candidate, Sensys 2022
References in corpus (17)
- Communication-Efficient Learning of Deep Networks from Decentralized Data
- Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
- Mobile Sensor Data Anonymization
- A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics
- FastDeepIoT: Towards Understanding and Optimizing Neural Network Execution Time on Mobile and Embedded Devices
- Thieves on Sesame Street! Model Extraction of BERT-based APIs
- HyperNetworks
- On Lightweight Privacy-Preserving Collaborative Learning for IoT Objects
- TIPRDC: Task-Independent Privacy-Respecting Data Crowdsourcing Framework for Deep Learning with Anonymized Intermediate Representations
- PPFL: Privacy-preserving Federated Learning with Trusted Execution Environments
- Honest-but-Curious Nets: Sensitive Attributes of Private Inputs Can Be Secretly Coded into the Classifiers' Outputs
- Adversarial Distillation of Bayesian Neural Network Posteriors
- HyperGAN: A Generative Model for Diverse, Performant Neural Networks
- Privacy Partitioning: Protecting User Data During the Deep Learning Inference Phase
- MEGEX: Data-Free Model Extraction Attack against Gradient-Based Explainable AI
- SCOTCH: An Efficient Secure Computation Framework for Secure Aggregation
- Application-driven Privacy-preserving Data Publishing with Correlated Attributes