From af8f642c0fe220a21c35b7426342a7fef3895c4b Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Wed, 22 Jul 2020 22:41:03 -0400 Subject: [PATCH 1/5] Add another MIA paper --- README.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index d504210..2bb40ae 100644 --- a/README.md +++ b/README.md @@ -24,6 +24,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Membership inference attack against differentially private deep learning model**](http://www.tdp.cat/issues16/tdp.a289a17.pdf) (Rahman et al., 2018) - [**Comprehensive privacy analysis of deep learning: Passive and active white-box inference attacks against centralized and federated learning.**](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8835245) (Nasr et al., 2019) ([code](https://github.com/privacytrustlab/ml_privacy_meter)) - [**Logan: Membership inference attacks against generative models.**](https://content.sciendo.com/downloadpdf/journals/popets/2019/1/article-p133.xml) (Hayes et al. 2019) ([code](https://github.com/jhayes14/gen_mem_inf)) +- [**Privacy Risks of Securing Machine Learning Models against Adversarial Examples**](https://arxiv.org/abs/1905.10291) (Song et al., 2019) ([code](https://github.com/inspire-group/privacy-vs-robustness)) - [**Evaluating differentially private machine learning in practice**](https://www.usenix.org/system/files/sec19-jayaraman.pdf) (Jayaraman and Evans, 2019) ([code](https://github.com/bargavj/EvaluatingDPML)) - [**Ml-leaks: Model and data independent membership inference attacks and defenses on machine learning models**](https://www.ndss-symposium.org/wp-content/uploads/2019/02/ndss2019_03A-1_Salem_paper.pdf) (Salem et al., 2019) ([code](https://github.com/AhmedSalem2/ML-Leaks)) - [**Privacy risks of securing machine learning models against adversarial examples**](https://dl.acm.org/doi/pdf/10.1145/3319535.3354211) (Song L. et al., 2019) ([code](https://github.com/inspire-group/privacy-vs-robustness)) @@ -93,4 +94,4 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Extraction of Complex DNN Models: Real Threat or Boogeyman?**](https://arxiv.org/pdf/1910.05429.pdf) (Atli et al., 2020) - [**Stealing Neural Networks via Timing Side Channels**](https://arxiv.org/pdf/1812.11720.pdf) (Duddu et al., 2019) - [**DeepSniffer: A DNN Model Extraction Framework Based on Learning Architectural Hints**](https://dl.acm.org/doi/pdf/10.1145/3373376.3378460) (Hu et al., 2020) -- [**CSI NN: Reverse Engineering of Neural Network Architectures Through Electromagnetic Side Channel**](https://www.usenix.org/system/files/sec19-batina.pdf) (Batina et al., 2019) \ No newline at end of file +- [**CSI NN: Reverse Engineering of Neural Network Architectures Through Electromagnetic Side Channel**](https://www.usenix.org/system/files/sec19-batina.pdf) (Batina et al., 2019) From 881c58c954410b091254b22121f0b68280560d5e Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Sat, 25 Jul 2020 10:41:31 -0400 Subject: [PATCH 2/5] Add "Stolen Memories:..." paper --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 2bb40ae..876f704 100644 --- a/README.md +++ b/README.md @@ -41,6 +41,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Modelling and Quantifying Membership Information Leakage in Machine Learning**](https://arxiv.org/pdf/2001.10648.pdf) (Farokhi and Kaafar, 2020) - [**Systematic Evaluation of Privacy Risks of Machine Learning Models**](https://arxiv.org/abs/2003.10595) (Song and Mittal, 2020) ([code](https://github.com/inspire-group/membership-inference-evaluation)) - [**Towards the Infeasibility of Membership Inference on Deep Models**](https://arxiv.org/pdf/2005.13702.pdf) (Rezaei and Liu, 2020) ([code](https://github.com/shrezaei/MI-Attack)) +- [**Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inference**](https://arxiv.org/abs/1906.11798) (Leino and Fredrikson, 2020) ## Reconstruction Reconstruction attacks cover also attacks known as *model inversion* and *attribute inference*. From 71f918e4fdb2ee8cf4ca7e1b155e3529a00bb499 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Sat, 25 Jul 2020 16:45:24 -0400 Subject: [PATCH 3/5] Add 1 reconstruction paper --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 876f704..44b6a92 100644 --- a/README.md +++ b/README.md @@ -68,6 +68,7 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning**](https://arxiv.org/pdf/1812.00535.pdf) (Wang et al., 2018) - [**Exploring Image Reconstruction Attack in Deep Learning Computation Offloading**](https://dl.acm.org/doi/pdf/10.1145/3325413.3329791) (Oh and Lee, 2019) - [**I Know What You See: Power Side-Channel Attack on Convolutional Neural Network Accelerators**](https://arxiv.org/pdf/1803.05847.pdf) (Wei et al., 2019) +- [**Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning**](https://arxiv.org/abs/1904.01067) (Salem et al., 2019) - [**Illuminating the Dark or how to recover what should not be seen in FE-based classifiers**](https://eprint.iacr.org/2018/1001) (Carpov et al., 2020) ## Property inference From 52af41afd7db30cc83e17d2d2d200327ff3ea3a8 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Sat, 25 Jul 2020 16:47:56 -0400 Subject: [PATCH 4/5] Add 1 extraction paper --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 44b6a92..3f71771 100644 --- a/README.md +++ b/README.md @@ -97,3 +97,4 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Stealing Neural Networks via Timing Side Channels**](https://arxiv.org/pdf/1812.11720.pdf) (Duddu et al., 2019) - [**DeepSniffer: A DNN Model Extraction Framework Based on Learning Architectural Hints**](https://dl.acm.org/doi/pdf/10.1145/3373376.3378460) (Hu et al., 2020) - [**CSI NN: Reverse Engineering of Neural Network Architectures Through Electromagnetic Side Channel**](https://www.usenix.org/system/files/sec19-batina.pdf) (Batina et al., 2019) +- [**Cache Telepathy: Leveraging Shared Resource Attacks to Learn DNN Architectures**](https://www.usenix.org/conference/usenixsecurity20/presentation/yan) (Yan et al., 2020) From 4a6da5dbd2346f5eb79de5dcaaa168fcc81fefe0 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Wed, 29 Jul 2020 23:37:08 -0400 Subject: [PATCH 5/5] Add new MI paper --- README.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 3f71771..2d8f23c 100644 --- a/README.md +++ b/README.md @@ -28,7 +28,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Evaluating differentially private machine learning in practice**](https://www.usenix.org/system/files/sec19-jayaraman.pdf) (Jayaraman and Evans, 2019) ([code](https://github.com/bargavj/EvaluatingDPML)) - [**Ml-leaks: Model and data independent membership inference attacks and defenses on machine learning models**](https://www.ndss-symposium.org/wp-content/uploads/2019/02/ndss2019_03A-1_Salem_paper.pdf) (Salem et al., 2019) ([code](https://github.com/AhmedSalem2/ML-Leaks)) - [**Privacy risks of securing machine learning models against adversarial examples**](https://dl.acm.org/doi/pdf/10.1145/3319535.3354211) (Song L. et al., 2019) ([code](https://github.com/inspire-group/privacy-vs-robustness)) -- [**White-box vs Black-box: Bayes Optimal Strategies for Membership Inference**](http://proceedings.mlr.press/v97/sablayrolles19a.html) (Sablayrolles etal., 2019) +- [**White-box vs Black-box: Bayes Optimal Strategies for Membership Inference**](http://proceedings.mlr.press/v97/sablayrolles19a.html) (Sablayrolles et al., 2019) - [**Privacy risks of explaining machine learning models**](https://arxiv.org/abs/1907.00164) (Shokri et al., 2019) - [**Demystifying membership inference attacks in machine learning as a service**](https://ieeexplore.ieee.org/abstract/document/8634878) (Truex et al., 2019) - [**Monte carlo and reconstruction membership inference attacks against generative models**](https://content.sciendo.com/view/journals/popets/2019/4/article-p232.xml) (Hilprecht et al., 2019) @@ -42,6 +42,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Systematic Evaluation of Privacy Risks of Machine Learning Models**](https://arxiv.org/abs/2003.10595) (Song and Mittal, 2020) ([code](https://github.com/inspire-group/membership-inference-evaluation)) - [**Towards the Infeasibility of Membership Inference on Deep Models**](https://arxiv.org/pdf/2005.13702.pdf) (Rezaei and Liu, 2020) ([code](https://github.com/shrezaei/MI-Attack)) - [**Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inference**](https://arxiv.org/abs/1906.11798) (Leino and Fredrikson, 2020) +- [**Label-Only Membership Inference Attacks**](https://arxiv.org/abs/2007.14321) (Choquette Choo et al., 2020) ## Reconstruction Reconstruction attacks cover also attacks known as *model inversion* and *attribute inference*.