Amalgam: A Framework for Obfuscated Neural Network Training on the Cloud
Not on the map. Why?
Abstract (arXiv)
Training a proprietary Neural Network (NN) model with a proprietary dataset on the cloud comes at the risk of exposing the model architecture and the dataset to the cloud service provider. To tackle this problem, in this paper, we present an NN obfuscation framework, called Amalgam, to train NN models in a privacy-preserving manner in existing cloud-based environments. Amalgam achieves that by augmenting NN models and the datasets to be used for training with well-calibrated noise to "hide" both the original model architectures and training datasets from the cloud. After training, Amalgam extracts the original models from the augmented models and returns them to users. Our evaluation results with different computer vision and natural language processing models and datasets demonstrate that Amalgam: (i) introduces modest overheads into the training process without impacting its correctness, and (ii) does not affect the model's accuracy. The prototype implementation is available at: https://github.com/SifatTaj/amalgam
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 60 references that are in the map. First posted on arXiv 2 Jun 2024.
Builds on
34 of 60 references.
References · 34
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2016Densely Connected Convolutional Networkscited by 6,691
- 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
- 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
- 2018CBAM: Convolutional Block Attention Modulecited by 2,463
- 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
- 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,618
22 more
- 2019A Comprehensive Survey on Transfer Learningcited by 1,154
- 2019Deep Leakage from Gradientscited by 1,004
- 2020iDLG: Improved Deep Leakage from Gradientscited by 367
- 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
- 2019Learning Deep Transformer Models for Machine Translationcited by 309
- 2021CrypTen: Secure Multi-Party Computation Meets Machine Learningcited by 131
- 2018A General Approach to Adding Differential Privacy to Iterative Training Procedurescited by 84
- 2019CrypTFlow: Secure TensorFlow Inferencecited by 61
- 2021Privacy-Preserving Machine Learning: Methods, Challenges and Directionscited by 58
- 2021Privacy-Preserving Machine Learning with Fully Homomorphic Encryption for Deep Neural Networkcited by 57
- 2019XONN: XNOR-based Oblivious Deep Neural Network Inferencecited by 51
- 2019QUOTIENT: Two-Party Secure Neural Network Training and Predictioncited by 41
- 2020Hermes Attack: Steal DNN Models with Lossless Inference Accuracycited by 41
- 2020Private Summation in the Multi-Message Shuffle Modelcited by 34
- 2023KBNet: Kernel Basis Network for Image Restorationcited by 33
- 2019Open DNN Box by Power Side-Channel Attackcited by 28
- 2020DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for deep neural networkscited by 23
- 2017BLENDER: Enabling Local Search with a Hybrid Differential Privacy Modelcited by 14
- 2020PrivacyFL: A simulator for privacy-preserving and secure federated learningcited by 12
- 2019TensorSCONE: A Secure TensorFlow Framework using Intel SGXcited by 8
- 2023ModelObfuscator: Obfuscating Model Information to Protect Deployed ML-based Systemscited by 7
- 2020A Privacy-Preserving Distributed Architecture for Deep-Learning-as-a-Servicecited by 3
Built on by
Not cited yet.