Privacy-Preserving Deep Learning: Revisited and Enhanced

Phong, Le Trieu; Aono, Yoshinori; Hayashi, Takuya; Wang, Lihua; Moriai, Shiho

doi:10.1007/978-981-10-5421-1_9

Cited by 117 publications

(46 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, Health Insurance Portability and Accountability Act (HIPAA)-compliant storage systems have paved the way for more stringent privacy preservation. Studies have explored systems that enable multiple entities to jointly train AI models without sharing their input data sets — sharing only the trained model 106,107 . Other efforts use a decentralized ‘federated’ learning approach 108 .…”

Section: Ai Challenges In Medical Imagingmentioning

confidence: 99%

Artificial intelligence in radiology

et al. 2018

View full text Add to dashboard Cite

Artificial intelligence (AI) algorithms, particularly deep learning, have demonstrated remarkable progress in image-recognition tasks. Methods ranging from convolutional neural networks to variational autoencoders have found myriad applications in the medical image analysis field, propelling it forward at a rapid pace. Historically, in radiology practice, trained physicians visually assessed medical images for the detection, characterization and monitoring of diseases. AI methods excel at automatically recognizing complex patterns in imaging data and providing quantitative, rather than qualitative, assessments of radiographic characteristics. In this O pinion article, we establish a general understanding of AI methods, particularly those pertaining to image-based tasks. We explore how these methods could impact multiple facets of radiology, with a general focus on applications in oncology, and demonstrate ways in which these methods are advancing the field. Finally, we discuss the challenges facing clinical implementation and provide our perspective on how the domain could be advanced.

show abstract

Section: Ai Challenges In Medical Imagingmentioning

confidence: 99%

Artificial intelligence in radiology

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Aono et al [49] show that, in the collaborative deep learning protocol of [52], an honest-but-curious server can partially recover participants' training inputs from their gradient updates under the (greatly simplified) assumption that the batch consists of a single input. Furthermore, the technique is evaluated only on MNIST where all class members are visually similar.…”

Section: Related Workmentioning

confidence: 99%

Exploiting Unintended Feature Leakage in Collaborative Learning

Melis

Song

Cristofaro

et al. 2019

2019 IEEE Symposium on Security and Privacy (SP)

1,193

878

View full text Add to dashboard Cite

Collaborative machine learning and related techniques such as federated learning allow multiple participants, each with his own training dataset, to build a joint model by training locally and periodically exchanging model updates. We demonstrate that these updates leak unintended information about participants' training data and develop passive and active inference attacks to exploit this leakage. First, we show that an adversarial participant can infer the presence of exact data points-for example, specific locations-in others' training data (i.e., membership inference). Then, we show how this adversary can infer properties that hold only for a subset of the training data and are independent of the properties that the joint model aims to capture. For example, he can infer when a specific person first appears in the photos used to train a binary gender classifier. We evaluate our attacks on a variety of tasks, datasets, and learning configurations, analyze their limitations, and discuss possible defenses.

show abstract

“…This approach does not require clients to send their local data to the central server, but several works show that the clients' model updates leak information about their local data [53], [113], [120]. To counter this, some works focus on secure aggregation techniques for distributed NNs, based on HE [93], [94] or MPC [22]. Although encrypting the gradient values prevents the leakage of parties' confidential data to the central server, these solutions do not account for potential leakage from the aggregate values themselves.…”

Section: Related Workmentioning

confidence: 99%

POSEIDON: Privacy-Preserving Federated Neural Network Learning

Sav¹,

Pyrgelis²,

Troncoso-Pastoriza³

et al. 2021

Proceedings 2021 Network and Distributed System Security Symposium

View full text Add to dashboard Cite

Recent cryptographic approaches for private distributed learning, e.g., [119], [42], not only have limited ML functionalities, i.e., regularized or generalized linear models, but also employ traditional encryption schemes that make them vulnerable to post-quantum attacks. This should be cautiously considered, as recent advances in quantum computing [47], [87], [105], [116], increase the need for deploying quantum-resilient cryptographic schemes that eliminate Abstract-In this paper, we address the problem of privacypreserving training and evaluation of neural networks in an N-party, federated learning setting. We propose a novel system, POSEIDON, the first of its kind in the regime of privacy-preserving neural network training. It employs multiparty lattice-based cryptography to preserve the confidentiality of the training data, the model, and the evaluation data, under a passive-adversary model and collusions between up to N − 1 parties. To efficiently execute the secure backpropagation algorithm for training neural networks, we provide a generic packing approach that enables Single Instruction, Multiple Data (SIMD) operations on encrypted data. We also introduce arbitrary linear transformations within the cryptographic bootstrapping operation, optimizing the costly cryptographic computations over the parties, and we define a constrained optimization problem for choosing the cryptographic parameters. Our experimental results show that POSEIDON achieves accuracy similar to centralized or decentralized non-private approaches and that its computation and communication overhead scales linearly with the number of parties. POSEIDON trains a 3-layer neural network on the MNIST dataset with 784 features and 60K samples distributed among 10 parties in less than 2 hours.

show abstract

Privacy-Preserving Deep Learning: Revisited and Enhanced

Cited by 117 publications

References 7 publications

Artificial intelligence in radiology

Artificial intelligence in radiology

Exploiting Unintended Feature Leakage in Collaborative Learning

POSEIDON: Privacy-Preserving Federated Neural Network Learning

Contact Info

Product

Resources

About