RQP-SGD: Differential Private Machine Learning through Noisy SGD and Randomized Quantization
This work addresses the need for efficient and secure machine learning on IoT devices, though it appears incremental as it builds on existing DP-SGD and quantization methods.
The paper tackles the problem of training machine learning models with quantized discrete weights for low-memory edge devices while preserving data privacy, by introducing RQP-SGD, which combines differentially private SGD with randomized quantization, and demonstrates its efficacy over deterministic quantization through experiments on two datasets.
The rise of IoT devices has prompted the demand for deploying machine learning at-the-edge with real-time, efficient, and secure data processing. In this context, implementing machine learning (ML) models with real-valued weight parameters can prove to be impractical particularly for large models, and there is a need to train models with quantized discrete weights. At the same time, these low-dimensional models also need to preserve privacy of the underlying dataset. In this work, we present RQP-SGD, a new approach for privacy-preserving quantization to train machine learning models for low-memory ML-at-the-edge. This approach combines differentially private stochastic gradient descent (DP-SGD) with randomized quantization, providing a measurable privacy guarantee in machine learning. In particular, we study the utility convergence of implementing RQP-SGD on ML tasks with convex objectives and quantization constraints and demonstrate its efficacy over deterministic quantization. Through experiments conducted on two datasets, we show the practical effectiveness of RQP-SGD.