POSEIDON: Privacy-Preserving Federated Neural Network Learning
It addresses privacy concerns for parties in federated learning, though it is incremental as it builds on existing cryptographic and federated learning methods.
The paper tackles the problem of privacy-preserving neural network training in a federated learning setting, proposing POSEIDON, a system that uses multiparty lattice-based cryptography to protect data and model confidentiality while achieving accuracy similar to non-private approaches and training a 3-layer network on MNIST with 10 parties in under 2 hours.
In this paper, we address the problem of privacy-preserving 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.