Geometry of Sensitivity: Twice Sampling and Hybrid Clipping in Differential Privacy with Optimal Gaussian Noise and Application to Deep Learning

Geometry of Sensitivity: Twice Sampling and Hybrid Clipping in Differential Privacy with Optimal Gaussian Noise and Application to Deep Learning

We study the fundamental problem of the construction of optimal randomization in Differential Privacy (DP). Depending on the clipping strategy or additional properties of the processing function, the corresponding sensitivity set theoretically determines the necessary randomization to produce the re...

Bibliografiset tiedot
Päätekijät:	Xiao, Hanshen, Wan, Jun, Devadas, Srinivas
Muut tekijät:	Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Aineistotyyppi:	Artikkeli
Kieli:	English
Julkaistu:	ACM\|Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security 2023
Linkit:	https://hdl.handle.net/1721.1/153139

Samankaltaisia teoksia

Deep Learning with Gaussian Differential Privacy
Tekijä: Zhiqi Bu, et al.
Julkaistu: (2020-07-01)

Formal Privacy Proof of Data Encoding: The Possibility and Impossibility of Learnable Encryption
Tekijä: Xiao, Hanshen, et al.
Julkaistu: (2025)

Local differential privacy in decentralized optimization
Tekijä: Xiao, Hanshen.
Julkaistu: (2020)

Enhancing Efficiency in Privacy-Preserving Federated Learning for Healthcare: Adaptive Gaussian Clipping With DFT Aggregator
Tekijä: Muhammad Ayat Hidayat, et al.
Julkaistu: (2024-01-01)

Adaptive selection method of differential privacy GAN gradient clipping thresholds
Tekijä: Peng GUO, et al.
Julkaistu: (2018-05-01)

Don’t simulate twice: one-shot sensitivity analyses via automatic differentiation
Tekijä: Quera-Bofarull, A, et al.
Julkaistu: (2023)

Fractional Milne-type inequalities for twice differentiable functions
Tekijä: Areej A. Almoneef, et al.
Julkaistu: (2024-06-01)

XRD: Scalable messaging system with cryptographic privacy
Tekijä: Kwon, Albert Hyukjae, et al.
Julkaistu: (2021)

Sampling-based robust control of autonomous systems with non-Gaussian noise
Tekijä: Badings, T, et al.
Julkaistu: (2022)

Twice Kissed /
Tekijä: Jackson, Lisa, 1952-, author 218751
Julkaistu: (2009)

Twice as Natural
Julkaistu: (2023-07-01)

Noise Variance Optimization in Differential Privacy: A Game-Theoretic Approach Through Per-Instance Differential Privacy
Tekijä: Sehyun Ryu, et al.
Julkaistu: (2024-01-01)

"To clip or not to clip. That is no question!"
Tekijä: Aznar, M, et al.
Julkaistu: (2017)

Resistor Geometry Comparison With Respect to Current Noise and Trim Sensitivity
Tekijä: W. Ulbrich
Julkaistu: (1977-01-01)

The Stochastic Stability of Internal HIV Models with Gaussian White Noise and Gaussian Colored Noise
Tekijä: Xiying Wang, et al.
Julkaistu: (2019-01-01)

Church numerals‚ twice!
Tekijä: Hinze, R
Julkaistu: (2002)

Key-Free Image Encryption Algorithm Based on Self-Triggered Gaussian Noise Sampling
Tekijä: Kai Gao, et al.
Julkaistu: (2024-01-01)

Privacy-preserving knowledge transfer for intrusion detection with federated deep autoencoding gaussian mixture model
Tekijä: Chen, Yang, et al.
Julkaistu: (2022)

Deep Inference for Covariance Estimation: Learning Gaussian Noise Models for State Estimation
Tekijä: Liu, Katherine, et al.
Julkaistu: (2020)

Detection of Gaussian noise and its level using deep convolutonal neural network
Tekijä: Joon, H.C., et al.
Julkaistu: (2017)

Detection of Gaussian noise and its level using deep convolutional neural network
Tekijä: Chuah, J.H., et al.
Julkaistu: (2017)

Clipping Noise in Visible Light Communication Systems with OFDM and PAPR Reduction
Tekijä: Hussien Alrakah, et al.
Julkaistu: (2024-07-01)

Sample-Efficient Proper PAC Learning with Approximate Differential Privacy
Tekijä: Ghazi, Badih, et al.
Julkaistu: (2022)

Provable ICA with Unknown Gaussian Noise, and Implications for Gaussian Mixtures and Autoencoders
Tekijä: Arora, Sanjeev, et al.
Julkaistu: (2017)

Non-Gaussian Noise in Superconducting Circuits
Tekijä: McCourt, Trevor Johnathan
Julkaistu: (2023)

Communication in the presence of additive gaussian noise
Julkaistu: (2004)

You only love twice /
Tekijä: 457089 Wilde, Lori
Julkaistu: (2006)

Twice upon a time /
Tekijä: Forster, Kate (Kate Louise), author
Julkaistu: (2014)

Twice is enough for dangerous eigenvalues
Tekijä: Horning, A, et al.
Julkaistu: (2022)

Twice upon a wedding /
Tekijä: Stone, Jean, 1948-, author
Julkaistu: (2005)

Measure twice, cut once /
Tekijä: 319488 Tolpin, Jim
Julkaistu: (1996)

Some generalization of integral inequalities for twice differentiable mappings involving fractional integrals
Tekijä: Sarikaya Mehmet Zeki, et al.
Julkaistu: (2015-12-01)

Bullen-type inequalities for twice-differentiable functions by using conformable fractional integrals
Tekijä: Fatih Hezenci, et al.
Julkaistu: (2024-04-01)

FL-AGN: A Privacy-Enhanced Federated Learning Method Based on Adaptive Gaussian Noise for Resisting Gradient Inference Attacks
Tekijä: Zhifu Huang, et al.
Julkaistu: (2024-01-01)

Twice clustering based hybrid model for short-term passenger flow forecasting
Tekijä: Sheng Wang, et al.
Julkaistu: (2024-11-01)

Adaptive Gaussian Wiener Filter for CT-Scan Images with Gaussian Noise Variance
Tekijä: Kai Liang Lew, et al.
Julkaistu: (2024-02-01)

Extended Object Tracking Based on Gaussian Process in Non-Gaussian Noise Environment
Tekijä: Lifan Sun, et al.
Julkaistu: (2025-01-01)

An Improved Gaussian Mixture CKF Algorithm under Non-Gaussian Observation Noise
Tekijä: Hongjian Wang, et al.
Julkaistu: (2016-01-01)

Privacy view and target of differential privacy
Tekijä: Jingyu JIA, et al.
Julkaistu: (2023-10-01)

Differential privacy budget optimization based on deep learning in IoT
Tekijä: Dan LUO, et al.
Julkaistu: (2022-06-01)