LGAICRJun 9, 2023

Differentially Private Sharpness-Aware Training

arXiv:2306.05651v118 citationsh-index: 12Has Code
Originality Incremental advance
AI Analysis

This work addresses the problem of reduced accuracy in differentially private training for deep learning practitioners, offering an incremental improvement over existing methods.

The paper tackles the performance degradation in differentially private deep learning by investigating sharpness and proposing a new sharpness-aware training method that improves model performance from scratch and fine-tuning, with experimental results showing concrete gains.

Training deep learning models with differential privacy (DP) results in a degradation of performance. The training dynamics of models with DP show a significant difference from standard training, whereas understanding the geometric properties of private learning remains largely unexplored. In this paper, we investigate sharpness, a key factor in achieving better generalization, in private learning. We show that flat minima can help reduce the negative effects of per-example gradient clipping and the addition of Gaussian noise. We then verify the effectiveness of Sharpness-Aware Minimization (SAM) for seeking flat minima in private learning. However, we also discover that SAM is detrimental to the privacy budget and computational time due to its two-step optimization. Thus, we propose a new sharpness-aware training method that mitigates the privacy-optimization trade-off. Our experimental results demonstrate that the proposed method improves the performance of deep learning models with DP from both scratch and fine-tuning. Code is available at https://github.com/jinseongP/DPSAT.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes