awesome-ml-privacy-attacks
github.com/stratosphereips/awesome-ml-privacy-attacks ↗An awesome list of papers on privacy attacks against machine learning
639
GitHub Stars
237
Curated Resources
4
Categories
5 hours ago
Last Refreshed
Membership inferenceReconstructionProperty inference / Distribution inferenceModel extraction
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me model extraction resources from awesome-ml-privacy-attacks"
Installation instructions →What's inside
Model extraction
- ACTIVETHIEF: Model Extraction Using Active Learning and Unannotated Public Data
- Amnesiac Machine Learning
- Analyzing Information Leakage of Updates to Natural Language Models
- Black-Box Ripper: Copying black-box models using generative evolutionary algorithms
- BODAME: Bilevel Optimization for Defense Against Model Extraction
- Bounding Information Leakage in Machine Learning
Membership inference
- ADePT: Auto-encoder based Differentially Private Text Transformation
- Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning
- Alleviating Privacy Attacks via Causal Learning
- An Extension of Fano's Inequality for Characterizing Model Susceptibility to Membership Inference Attacks
- Auditing Data Provenance in Text-Generation Models
- Bootstrap Aggregation for Point-based Generalized Membership Inference Attacks
Resources
- Adversarial Robustness Toolbox (ART)
- An Overview of Privacy in Machine Learning
- A Review of Confidentiality Threats Against Embedded Neural Network Models
- A Survey of Privacy Attacks in Machine Learning
- CypherCat (archive-only)
- Federated Learning Attacks Revisited: A Critical Discussion of Gaps,Assumptions, and Evaluation Setups
Reconstruction
- A Framework for Evaluating Gradient Leakage Attacks in Federated Learning
- A Linear Reconstruction Approach for Attribute Inference Attacks against Synthetic Data
- A methodology for formalizing model-inversion attacks
- Analysis and Utilization of Hidden Information in Model Inversion Attacks
- An Attack-Based Evaluation Method for Differentially Private Learning Against Model Inversion Attack
- Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning
Property inference / Distribution inference
- Dissecting Distribution Inference
- Exploiting unintended feature leakage in collaborative learning
- Formalizing and Estimating Distribution Inference Risks
- Hacking smart machines with smarter ones: How to extract meaningful data from machine learning classifiers
- Honest-but-Curious Nets: Sensitive Attributes of Private Inputs can be Secretly Coded into the Entropy of Classifiers' Outputs
- Overlearning Reveals Sensitive Attributes
Showing a sample of 237 resources. View the full list on GitHub →