On the Privacy Risks of Cell-Based NAS Architectures

Abstract

Existing studies on neural architecture search (NAS) mainly focus on efficiently and effectively searching for network architectures with better performance. Little progress has been made to systematically understand if the NAS-searched architectures are robust to privacy attacks while abundant work has already shown that human-designed architectures are prone to privacy attacks. In this paper, we fill this gap and systematically measure the privacy risks of NAS architectures. Leveraging the insights from our measurement study, we further explore the cell patterns of cell-based NAS architectures and evaluate how the cell patterns affect the privacy risks of NAS-searched architectures. Through extensive experiments, we shed light on how to design robust NAS architectures against privacy attacks, and also offer a general methodology to understand the hidden correlation between the NAS-searched architectures and other privacy risks.

arXiv commentAccepted by CCS 2022

References in corpus (31)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• XGBoost: A Scalable Tree Boosting System
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Neural Architecture Search with Reinforcement Learning
• Language Models are Few-Shot Learners
• Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
• Efficient Neural Architecture Search via Parameter Sharing
• Designing Neural Network Architectures using Reinforcement Learning
• GAN-Leaks: A Taxonomy of Membership Inference Attacks against Generative Models
• One-Shot Neural Architecture Search via Self-Evaluated Template Network
• Path-Level Network Transformation for Efficient Architecture Search
• Depth-Width Tradeoffs in Approximating Natural Functions with Neural Networks
• Label-Only Membership Inference Attacks
• Evaluating Differentially Private Machine Learning in Practice
• NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search
• Stabilizing Differentiable Architecture Search via Perturbation-based Regularization
• Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective
• Node-Level Membership Inference Attacks Against Graph Neural Networks
• ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models
• Rethinking Architecture Selection in Differentiable NAS
• NAS-FCOS: Fast Neural Architecture Search for Object Detection
• Membership Leakage in Label-Only Exposures
• Neural Architecture Dilation for Adversarial Robustness
• Understanding Architectures Learnt by Cell-based Neural Architecture Search
• Membership Inference Attacks Against Recommender Systems
• On Redundancy and Diversity in Cell-based Neural Architecture Search
• Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets
• Generalization Guarantees for Neural Architecture Search with Train-Validation Split
• DOTS: Decoupling Operation and Topology in Differentiable Architecture Search

Cited by in corpus (2)

• ERASER: Machine Unlearning in MLaaS via an Inference Serving-Aware Approach
• ORL-AUDITOR: Dataset Auditing in Offline Deep Reinforcement Learning