# Training with Differential Privacy: A Gradient-Preserving Noise Reduction Approach with Provable Security

arXiv: https://arxiv.org/abs/2409.11663 · HTML: https://mlsanity.com/paper/2409.11663/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 59 references that are in the map.
- First posted on arXiv 18 Sep 2024.

## Builds on (most cited first, 30 of 30)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Renyi Differential Privacy](https://mlsanity.com/paper/1702.07476/) (2017)
- [Opacus: User-Friendly Differential Privacy Library in PyTorch](https://mlsanity.com/paper/2109.12298/) (2021)
- [Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data](https://mlsanity.com/paper/1610.05755/) (2016)
- [RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response](https://mlsanity.com/paper/1407.6981/) (2014)
- [Scalable Private Learning with PATE](https://mlsanity.com/paper/1802.08908/) (2018)
- [Differentially Private Learning Needs Better Features (or Much More Data)](https://mlsanity.com/paper/2011.11660/) (2020)
- [Tempered Sigmoid Activations for Deep Learning with Differential Privacy](https://mlsanity.com/paper/2007.14191/) (2020)
- [Large Scale Private Learning via Low-rank Reparametrization](https://mlsanity.com/paper/2106.09352/) (2021)
- [Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning](https://mlsanity.com/paper/2102.12677/) (2021)
- [Differentially Private Model Publishing for Deep Learning](https://mlsanity.com/paper/1904.02200/) (2019)
- [Privacy in Deep Learning: A Survey](https://mlsanity.com/paper/2004.12254/) (2020)
- [DP-Forward: Fine-tuning and Inference on Language Models with Differential Privacy in Forward Pass](https://mlsanity.com/paper/2309.06746/) (2023)
- [CirCNN: Accelerating and Compressing Deep Neural Networks Using Block-CirculantWeight Matrices](https://mlsanity.com/paper/1708.08917/) (2017)
- [Private Adaptive Gradient Methods for Convex Optimization](https://mlsanity.com/paper/2106.13756/) (2021)
- [GAP: Differentially Private Graph Neural Networks with Aggregation Perturbation](https://mlsanity.com/paper/2203.00949/) (2022)
- [Improving Deep Learning with Differential Privacy using Gradient Encoding and Denoising](https://mlsanity.com/paper/2007.11524/) (2020)
- [Exploring the Vulnerability of Deep Neural Networks: A Study of Parameter Corruption](https://mlsanity.com/paper/2006.05620/) (2020)
- [Privacy-preserving Adversarial Facial Features](https://mlsanity.com/paper/2305.05391/) (2023)
- [DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy](https://mlsanity.com/paper/2110.08557/) (2021)
- [Differentially Private Set Union](https://mlsanity.com/paper/2002.09745/) (2020)
- [Unraveling the Connections between Privacy and Certified Robustness in Federated Learning Against Poisoning Attacks](https://mlsanity.com/paper/2209.04030/) (2022)
- [Preserving Differential Privacy in Convolutional Deep Belief Networks](https://mlsanity.com/paper/1706.08839/) (2017)
- [Marginal Release Under Local Differential Privacy](https://mlsanity.com/paper/1711.02952/) (2017)
- [Cross-silo Federated Learning with Record-level Personalized Differential Privacy](https://mlsanity.com/paper/2401.16251/) (2024)
- [DPAdapter: Improving Differentially Private Deep Learning through Noise Tolerance Pre-training](https://mlsanity.com/paper/2403.02571/) (2024)
- [FFT-Based Deep Learning Deployment in Embedded Systems](https://mlsanity.com/paper/1712.04910/) (2017)
- [Spectral-DP: Differentially Private Deep Learning through Spectral Perturbation and Filtering](https://mlsanity.com/paper/2307.13231/) (2023)
- [Towards Differentially Private Truth Discovery for Crowd Sensing Systems](https://mlsanity.com/paper/1810.04760/) (2018)

## Abstract (arXiv)

Deep learning models have been extensively adopted in various regions due to their ability to represent hierarchical features, which highly rely on the training set and procedures. Thus, protecting the training process and deep learning algorithms is paramount in privacy preservation. Although Differential Privacy (DP) as a powerful cryptographic primitive has achieved satisfying results in deep learning training, the existing schemes still fall short in preserving model utility, i.e., they either invoke a high noise scale or inevitably harm the original gradients. To address the above issues, in this paper, we present a more robust and provably secure approach for differentially private training called GReDP. Specifically, we compute the model gradients in the frequency domain and adopt a new approach to reduce the noise level. Unlike previous work, our GReDP only requires half of the noise scale compared to DPSGD [1] while keeping all the gradient information intact. We present a detailed analysis of our method both theoretically and empirically. The experimental results show that our GReDP works consistently better than the baselines on all models and training settings.