paper-with-me

Papers

Scalable Differential Privacy with Certified Robustness in Adversarial Learning

2020-01-01 · ICML 2020 1 · Hai Phan, My T. Thai, Han Hu, Ruoming Jin, Tong Sun, Dejing Dou

In this paper, we aim to develop a scalable algorithm to preserve differential privacy (DP) in adversarial learning for deep neural networks (DNNs), with certified robustness to adversarial examples. By leveraging the sequential composition theory in DP, we randomize both input and latent spaces to strengthen our certified robustness bounds. To address the trade-off among model utility, privacy loss, and robustness, we design an original adversarial objective function, based on the post-processing property in DP, to tighten the sensitivity of our model. A new stochastic batch training is proposed to apply our mechanism on large DNNs and datasets, by bypassing the vanilla iterative batch-by-batch training in DP DNNs. An end-to-end theoretical analysis and evaluations show that our mechanism notably improves the robustness and scalability of DP DNNs.

📄 PDF Abstract BibTeX

Code (1)

haiphanNJIT/StoBatch 공식 구현 tf

Similar Papers 제목 키워드 기반

Augment then Smooth: Reconciling Differential Privacy with Certified Robustness

2023-06-14 · Jiapeng Wu, Atiyeh Ashari Ghomi, David Glukhov, Jesse C. Cresswell 외

Machine learning models are susceptible to a variety of attacks that can erode trust, including attacks against the privacy of training data, and adversarial examples that jeopardize model accuracy. Differential privacy …

Unraveling the Connections between Privacy and Certified Robustness in Federated Learning Against Poisoning Attacks

2022-09-08 · Chulin Xie, Yunhui Long, Pin-Yu Chen, Qinbin Li 외

Federated learning (FL) provides an efficient paradigm to jointly train a global model leveraging data from distributed users. As local training data comes from different users who may not be trustworthy, several studies…

Federated Learning

Certified Robustness to Adversarial Examples with Differential Privacy

2018-02-09 · Mathias Lecuyer, Vaggelis Atlidakis, Roxana Geambasu, Daniel Hsu 외

Adversarial examples that fool machine learning models, particularly deep neural networks, have been a topic of intense research interest, with attacks and defenses being developed in a tight back-and-forth. Most past de…

Certified Robustness for Free in Differentially Private Federated Learning

2021-09-29 · Chulin Xie, Yunhui Long, Pin-Yu Chen, Krishnaram Kenthapadi 외

Federated learning (FL) provides an efficient training paradigm to jointly train a global model leveraging data from distributed users. As the local training data comes from different users who may not be trustworthy, s…

Federated Learning

Certified Robustness to Word Substitution Attack with Differential Privacy

2021-06-01 · NAACL 2021 4 · Wenjie Wang, Pengfei Tang, Jian Lou, Li Xiong

The robustness and security of natural language processing (NLP) models are significantly important in real-world applications. In the context of text classification tasks, adversarial examples can be designed by substit…

Adversarial RobustnessClassificationtext-classificationText Classification