From 55494f413f40945ce61fe1f72f732eb146d5c7e2 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain <17119557+suhacker1@users.noreply.github.com> Date: Fri, 9 Jul 2021 02:26:48 -0400 Subject: [PATCH 1/6] Add more papers --- README.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/README.md b/README.md index 8143162..9101e77 100644 --- a/README.md +++ b/README.md @@ -20,6 +20,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Privacy and Security Issues in Deep Learning: A Survey**](https://ieeexplore.ieee.org/abstract/document/9294026) (Liu et al., 2021) - [**ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models**](https://arxiv.org/abs/2102.02551) (Liu et al., 2021) - [**Membership Inference Attacks on Machine Learning: A Survey**](https://arxiv.org/abs/2103.07853) (Hu et al., 2021) +- [**Survey: Leakage and Privacy at Inference Time**](https://arxiv.org/abs/2107.01614) (Jegorova et al., 2021) # Privacy Testing Tools - [**PrivacyRaven**](https://github.com/trailofbits/PrivacyRaven) (Trail of Bits) @@ -87,6 +88,8 @@ This repository contains a curated list of papers related to privacy attacks aga - [**The Influence of Dropout on Membership Inference in Differentially Private Models**](https://arxiv.org/abs/2103.09008) (Galinkin, 2021) - [**Membership Inference Attack Susceptibility of Clinical Language Models**](https://arxiv.org/abs/2104.08305) (Jagannatha et al., 2021) - [**Membership Inference Attacks on Knowledge Graphs**](https://arxiv.org/abs/2104.08273) (Wang & Sun, 2021) +- [**When Does Data Augmentation Help With Membership Inference Attacks?**](http://proceedings.mlr.press/v139/kaya21a.html) (Kaya and Dumitras, 2021) +- [**The Influence of Training Parameters and Architectural Choices on the Vulnerability of Neural Networks to Membership Inference Attacks**](https://www.mi.fu-berlin.de/inf/groups/ag-idm/theseses/2021_oussama_bouanani_bsc_thesis.pdf) (Bouanani, 2021) ## Reconstruction @@ -140,6 +143,7 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Practical Defences Against Model Inversion Attacks for Split Neural Networks**](https://arxiv.org/abs/2104.05743) (Titcombe et al., 2021) - [**R-GAP: Recursive Gradient Attack on Privacy**](https://arxiv.org/abs/2010.07733) (Zhu and Blaschko, 2021) - [**Exploiting Explanations for Model Inversion Attacks**](https://arxiv.org/abs/2104.12669) (Zhao et al., 2021) +- [**SAFELearn: Secure Aggregation for private FEderated Learning**](https://encrypto.de/papers/FMMMMNRSSYZ21.pdf) (Fereidooni et al., 2021) ## Property inference @@ -217,3 +221,5 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Coded Machine Unlearning**](https://arxiv.org/abs/2012.15721) (Aldaghri et al., 2020) - [**Unlearnable Examples: Making Personal Data Unexploitable**](https://arxiv.org/abs/2101.04898) (Huang et al., 2021) - [**Measuring Data Leakage in Machine-Learning Models with Fisher Information**](https://arxiv.org/abs/2102.11673) (Hannun et al., 2021) +- [**Teacher Model Fingerprinting Attacks Against Transfer Learning**](https://arxiv.org/abs/2106.12478) (Chen et al, 2021) +- [**Bounding Information Leakage in Machine Learning**](https://arxiv.org/abs/2105.03875) (Del Grosso et al., 2021) From b78af51ef3732e3cd674b1a21ec937145d7ff3e1 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain <17119557+suhacker1@users.noreply.github.com> Date: Fri, 9 Jul 2021 02:38:05 -0400 Subject: [PATCH 2/6] Update README.md --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index 9101e77..b893893 100644 --- a/README.md +++ b/README.md @@ -223,3 +223,5 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Measuring Data Leakage in Machine-Learning Models with Fisher Information**](https://arxiv.org/abs/2102.11673) (Hannun et al., 2021) - [**Teacher Model Fingerprinting Attacks Against Transfer Learning**](https://arxiv.org/abs/2106.12478) (Chen et al, 2021) - [**Bounding Information Leakage in Machine Learning**](https://arxiv.org/abs/2105.03875) (Del Grosso et al., 2021) +- [**RoFL: Attestable Robustness for Secure Federated Learning**](https://arxiv.org/abs/2107.03311) +- From 0b84ebd0b9c3bd6c1723aaaacbd1ba521ff4d1d9 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain <17119557+suhacker1@users.noreply.github.com> Date: Fri, 9 Jul 2021 02:38:24 -0400 Subject: [PATCH 3/6] Update README.md --- README.md | 1 - 1 file changed, 1 deletion(-) diff --git a/README.md b/README.md index b893893..8175234 100644 --- a/README.md +++ b/README.md @@ -224,4 +224,3 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Teacher Model Fingerprinting Attacks Against Transfer Learning**](https://arxiv.org/abs/2106.12478) (Chen et al, 2021) - [**Bounding Information Leakage in Machine Learning**](https://arxiv.org/abs/2105.03875) (Del Grosso et al., 2021) - [**RoFL: Attestable Robustness for Secure Federated Learning**](https://arxiv.org/abs/2107.03311) -- From 8ba0714532f179a8e6ec7b0bbf94198b01b0360a Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain <17119557+suhacker1@users.noreply.github.com> Date: Fri, 9 Jul 2021 03:05:11 -0400 Subject: [PATCH 4/6] Update README.md --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index 8175234..0baa279 100644 --- a/README.md +++ b/README.md @@ -90,6 +90,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**Membership Inference Attacks on Knowledge Graphs**](https://arxiv.org/abs/2104.08273) (Wang & Sun, 2021) - [**When Does Data Augmentation Help With Membership Inference Attacks?**](http://proceedings.mlr.press/v139/kaya21a.html) (Kaya and Dumitras, 2021) - [**The Influence of Training Parameters and Architectural Choices on the Vulnerability of Neural Networks to Membership Inference Attacks**](https://www.mi.fu-berlin.de/inf/groups/ag-idm/theseses/2021_oussama_bouanani_bsc_thesis.pdf) (Bouanani, 2021) +- [**Membership Inference on Word Embedding and Beyond**](https://arxiv.org/abs/2106.11384) (Mahloujifar et al., 2021) ## Reconstruction @@ -144,6 +145,7 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**R-GAP: Recursive Gradient Attack on Privacy**](https://arxiv.org/abs/2010.07733) (Zhu and Blaschko, 2021) - [**Exploiting Explanations for Model Inversion Attacks**](https://arxiv.org/abs/2104.12669) (Zhao et al., 2021) - [**SAFELearn: Secure Aggregation for private FEderated Learning**](https://encrypto.de/papers/FMMMMNRSSYZ21.pdf) (Fereidooni et al., 2021) +- [**Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?**](https://arxiv.org/abs/2104.07762) (Lehman et al., 2021) ## Property inference From ea8bcee6ea95d1a09f6fe9a3a41fda66e971cc1b Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain <17119557+suhacker1@users.noreply.github.com> Date: Fri, 9 Jul 2021 03:07:03 -0400 Subject: [PATCH 5/6] Update README.md --- README.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 0baa279..e788379 100644 --- a/README.md +++ b/README.md @@ -130,7 +130,6 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**SAPAG: A Self-Adaptive Privacy Attack From Gradients**](https://arxiv.org/abs/2009.06228) (Wang et al., 2020) - [**Theory-Oriented Deep Leakage from Gradients via Linear Equation Solver**](https://arxiv.org/abs/2010.13356) (Pan et al., 2020) - [**Improved Techniques for Model Inversion Attacks**](https://arxiv.org/abs/2010.04092) (Chen et al., 2020) -- [**KART: Privacy Leakage Framework of Language Models Pre-trained with Clinical Records**](https://arxiv.org/abs/2101.00036) (Nakamura et al., 2020) - [**Black-box Model Inversion Attribute Inference Attacks on Classification Models**](https://arxiv.org/abs/2012.03404) (Mehnaz et al., 2020) - [**Deep Face Recognizer Privacy Attack: Model Inversion Initialization by a Deep Generative Adversarial Data Space Discriminator**](https://ieeexplore.ieee.org/abstract/document/9306253?casa_token=H78uIRJ2smYAAAAA:iQiA_5d2a2mbH4oBF9EZwSjakAz3Muq3ZOkNDBkK_fLq19PEMGEvpipyli7d9SGKESglqIb9Ug) (Khosravy et al., 2020) - [**MixCon: Adjusting the Separability of Data Representations for Harder Data Recovery**](https://arxiv.org/abs/2010.11463) (Li et al., 2020) @@ -146,6 +145,8 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**Exploiting Explanations for Model Inversion Attacks**](https://arxiv.org/abs/2104.12669) (Zhao et al., 2021) - [**SAFELearn: Secure Aggregation for private FEderated Learning**](https://encrypto.de/papers/FMMMMNRSSYZ21.pdf) (Fereidooni et al., 2021) - [**Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?**](https://arxiv.org/abs/2104.07762) (Lehman et al., 2021) +- [**Training Data Leakage Analysis in Language Models**](https://arxiv.org/abs/2101.05405) (Inan et al., 2021) +- ## Property inference From cb64b4efa19a3ee2d6e258b5cf82128b8d200539 Mon Sep 17 00:00:00 2001 From: Suha Sabi Hussain Date: Fri, 27 Aug 2021 15:36:12 -0400 Subject: [PATCH 6/6] Update README.md --- README.md | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index e788379..b104f57 100644 --- a/README.md +++ b/README.md @@ -21,6 +21,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models**](https://arxiv.org/abs/2102.02551) (Liu et al., 2021) - [**Membership Inference Attacks on Machine Learning: A Survey**](https://arxiv.org/abs/2103.07853) (Hu et al., 2021) - [**Survey: Leakage and Privacy at Inference Time**](https://arxiv.org/abs/2107.01614) (Jegorova et al., 2021) +- [**A Review of Confidentiality Threats Against Embedded Neural Network Models**](https://arxiv.org/abs/2105.01401) (Joud et al., 2021) # Privacy Testing Tools - [**PrivacyRaven**](https://github.com/trailofbits/PrivacyRaven) (Trail of Bits) @@ -91,6 +92,7 @@ This repository contains a curated list of papers related to privacy attacks aga - [**When Does Data Augmentation Help With Membership Inference Attacks?**](http://proceedings.mlr.press/v139/kaya21a.html) (Kaya and Dumitras, 2021) - [**The Influence of Training Parameters and Architectural Choices on the Vulnerability of Neural Networks to Membership Inference Attacks**](https://www.mi.fu-berlin.de/inf/groups/ag-idm/theseses/2021_oussama_bouanani_bsc_thesis.pdf) (Bouanani, 2021) - [**Membership Inference on Word Embedding and Beyond**](https://arxiv.org/abs/2106.11384) (Mahloujifar et al., 2021) +- [**TableGAN-MCA: Evaluating Membership Collisions of GAN-Synthesized Tabular Data Releasing**](https://arxiv.org/abs/2107.13190) (Hu et al., 2021) ## Reconstruction @@ -146,7 +148,12 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**SAFELearn: Secure Aggregation for private FEderated Learning**](https://encrypto.de/papers/FMMMMNRSSYZ21.pdf) (Fereidooni et al., 2021) - [**Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?**](https://arxiv.org/abs/2104.07762) (Lehman et al., 2021) - [**Training Data Leakage Analysis in Language Models**](https://arxiv.org/abs/2101.05405) (Inan et al., 2021) -- +- [**Exploiting Explanations for Model Inversion Attacks**](https://arxiv.org/abs/2104.12669) (Zhao et al., 2021) +- [**Model Fragmentation, Shuffle and Aggregation to Mitigate Model Inversion in Federated Learning**](https://ieeexplore.ieee.org/abstract/document/9478813?casa_token=047c6zFuwm4AAAAA:h6qWPCm6WXUbtVgk1iATPshiPMfvGEp6lVUrblEm8P2tRX4OIDEDpnzICVwYveoENEnH6Ig-yg) (Masude et al., 2021) +- [**PRECODE - A Generic Model Extension to Prevent Deep Gradient Leakage**](https://arxiv.org/abs/2108.04725) (Scheliga et al., 2021) +- [**On the Importance of Encrypting Deep Features**](https://arxiv.org/abs/2108.07147) (Ni et al., 2021) +- [**Defending Against Model Inversion Attack by Adversarial Examples**](https://www.cs.hku.hk/data/techreps/document/TR-2021-03.pdf) (Wen et al., 2021) +- [****]() (et al., 2021) ## Property inference @@ -208,6 +215,11 @@ Reconstruction attacks cover also attacks known as *model inversion* and *attrib - [**BODAME: Bilevel Optimization for Defense Against Model Extraction**](https://arxiv.org/abs/2103.06797) (Mori et al., 2021) - [**Dataset Inference: Ownership Resolution in Machine Learning**](https://openreview.net/forum?id=hvdKKV2yt7T) (Maini et al., 2021) - [**Good Artists Copy, Great Artists Steal: Model Extraction Attacks Against Image Translation Generative Adversarial Networks**](https://arxiv.org/abs/2104.12623) (Szyller et al., 2021) +- [**Towards Characterizing Model Extraction Queries and How to Detect Them**](https://www2.eecs.berkeley.edu/Pubs/TechRpts/2021/EECS-2021-126.pdf) (Zhang et al., 2021) +- [**Hardness of Samples Is All You Need: Protecting Deep Learning Models Using Hardness of Samples**](https://arxiv.org/abs/2106.11424) (Sadeghzadeh et al., 2021) +- [**Stateful Detection of Model Extraction Attacks**](https://arxiv.org/abs/2107.05166) (Pal et al., 2021) +- [**MEGEX: Data-Free Model Extraction Attack against Gradient-Based Explainable AI**](https://arxiv.org/abs/2107.08909) (Miura et al., 2021) +- [**INVERSENET: Augmenting Model Extraction Attacks with Training Data Inversion**](https://www.ijcai.org/proceedings/2021/0336.pdf) (Gong et al., 2021) # Other