Efficient Deep Learning on Multi-Source Private Data

Abstract

Machine learning models benefit from large and diverse datasets. Using such datasets, however, often requires trusting a centralized data aggregator. For sensitive applications like healthcare and finance this is undesirable as it could compromise patient privacy or divulge trade secrets. Recent advances in secure and privacy-preserving computation, including trusted hardware enclaves and differential privacy, offer a way for mutually distrusting parties to efficiently train a machine learning model without revealing the training data. In this work, we introduce Myelin, a deep learning framework which combines these privacy-preservation primitives, and use it to establish a baseline level of performance for fully private machine learning.

References in corpus (5)

• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Deep Learning with Differential Privacy
• Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware
• Chiron: Privacy-preserving Machine Learning as a Service
• Efficient Per-Example Gradient Computations

Cited by in corpus (31)

• DarkneTZ: Towards Model Privacy at the Edge using Trusted Execution Environments
• Split learning for health: Distributed deep learning without sharing raw patient data
• Blockchain-based Federated Learning with Secure Aggregation in Trusted Execution Environment for Internet-of-Things
• A Roadmap Towards Resilient Internet of Things for Cyber-Physical Systems
• A Critical Review on the Use (and Misuse) of Differential Privacy in Machine Learning
• Evaluating Differentially Private Machine Learning in Practice
• Machine Learning with Confidential Computing: A Systematization of Knowledge
• Leakage of Dataset Properties in Multi-Party Machine Learning
• Privado: Practical and Secure DNN Inference with Enclaves
• Opacus: User-Friendly Differential Privacy Library in PyTorch
• Differentially private cross-silo federated learning
• Confidential Inference via Ternary Model Partitioning
• An Overview of Privacy in Machine Learning
• DarKnight: A Data Privacy Scheme for Training and Inference of Deep Neural Networks
• Confidential Machine Learning Computation in Untrusted Environments: A Systems Security Perspective
• Towards Security Threats of Deep Learning Systems: A Survey
• Key Protected Classification for Collaborative Learning
• Privacy-Preserving Multi-Party Contextual Bandits
• GOAT: GPU Outsourcing of Deep Learning Training With Asynchronous Probabilistic Integrity Verification Inside Trusted Execution Environment
• Mitigating Leakage in Federated Learning with Trusted Hardware
• SESAME: Software defined Enclaves to Secure Inference Accelerators with Multi-tenant Execution
• Federated learning with differential privacy and an untrusted aggregator
• Citadel: Protecting Data Privacy and Model Confidentiality for Collaborative Learning with SGX
• Separation of Powers in Federated Learning
• Reaching Data Confidentiality and Model Accountability on the CalTrain
• Confidential Attestation: Efficient in-Enclave Verification of Privacy Policy Compliance
• An Extension of Fano's Inequality for Characterizing Model Susceptibility to Membership Inference Attacks
• Sharing in a Trustless World: Privacy-Preserving Data Analytics with Potentially Cheating Participants
• Accelerating 2PC-based ML with Limited Trusted Hardware
• Secure Machine Learning in the Cloud Using One Way Scrambling by Deconvolution
• Privacy and Integrity Preserving Training Using Trusted Hardware