arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2011.14572cs.LGcs.CRmath.OCstat.ML
收藏

梯度稀疏化可以提高差分私有凸机器学习的性能

Gradient Sparsification Can Improve Performance of Differentially-Private Convex Machine Learning

The University of Melbourne(墨尔本大学)

Farhad Farokhi

摘要:

我们使用梯度稀疏化来减少差分隐私噪声对私有机器学习模型性能的不利影响。为此,我们采用压缩感知和加性拉普拉斯噪声来评估差分私有梯度。噪声隐私保护梯度用于执行随机梯度下降以训练机器学习模型。通过将最小梯度项设置为零来实现的稀疏化可以降低训练算法的收敛速度。然而,通过稀疏化和压缩感知,可以降低所传递的梯度的维数和加性噪声的幅度。这些效应之间的相互作用决定了梯度稀疏化是否提高了差分私有机器学习模型的性能。我们在论文中对此进行了分析研究。我们证明,对于小的隐私预算,压缩可以提高隐私保护机器学习模型的性能。然而,对于大的隐私预算,压缩不一定提高性能。直观地说,这是因为隐私保护噪声的影响在大隐私预算制度中是最小的,因此梯度稀疏化的改进不能补偿其较慢的收敛。

英文摘要:

We use gradient sparsification to reduce the adverse effect of differential privacy noise on performance of private machine learning models. To this aim, we employ compressed sensing and additive Laplace noise to evaluate differentially-private gradients. Noisy privacy-preserving gradients are used to perform stochastic gradient descent for training machine learning models. Sparsification, achieved by setting the smallest gradient entries to zero, can reduce the convergence speed of the training algorithm. However, by sparsification and compressed sensing, the dimension of communicated gradient and the magnitude of additive noise can be reduced. The interplay between these effects determines whether gradient sparsification improves the performance of differentially-private machine learning models. We investigate this analytically in the paper. We prove that, for small privacy budgets, compression can improve performance of privacy-preserving machine learning models. However, for large privacy budgets, compression does not necessarily improve the performance. Intuitively, this is because the effect of privacy-preserving noise is minimal in large privacy budget regime and thus improvements from gradient sparsification cannot compensate for its slower convergence.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑