Steal-ML
Model attack tools
An implementation of extraction attacks against Machine Learning models offered by Cloud-based services
Model extraction attacks on Machine-Learning-as-a-Service platforms.
344 stars
18 watching
96 forks
Language: Python
last commit: almost 4 years ago Related projects:
| Repository | Description | Stars |
|---|---|---|
| ftramer/lm_memorization | A tool to extract memorized content from large language models like GPT-2 by analyzing their training data | 175 |
 freedomintelligence/mllm-bench | Evaluates and compares the performance of multimodal large language models on various tasks | 55 |
 dask/dask-ml | A Python library for scalable machine learning using Dask alongside popular ML libraries | 902 |
 jphall663/interpretable_machine_learning_with_python | Teaching software developers how to build transparent and explainable machine learning models using Python | 673 |
 lhfowl/robbing_the_fed | This implementation allows an attacker to directly obtain user data from federated learning gradient updates by modifying the shared model architecture. | 23 |
 princeton-sysml/film | Recovering private text in federated learning of language models by identifying and reconstructing sensitive client data from model gradients | 57 |
 inspire-group/modelpoisoning | An implementation of model poisoning attacks in federated learning | 146 |
 aws-samples/machine-learning-samples | A collection of sample applications demonstrating the use of Amazon Machine Learning API for various machine learning tasks | 872 |
 mitre/advmlthreatmatrix | A framework to help security analysts understand and prepare for adversarial machine learning attacks on AI systems | 1,050 |
 mlcommons/inference | Measures the performance of deep learning models in various deployment scenarios. | 1,236 |
 tensorflow/model-analysis | Evaluates and visualizes the performance of machine learning models. | 1,258 |
 eightbec/fastapi-ml-skeleton | A FastAPI-based framework for serving machine learning models in production-ready applications | 394 |
 rek7/mxtract | Analyzes and dumps memory to extract sensitive information from running processes | 582 |
 m0rtem/cloudfail | Reconnaissance tool that gathers information about a Cloudflare-protected target to discover its server location using misconfigured DNS and old database records. | 2,237 |
 digitalglobe/mltools | Tools for building machine learning solutions on satellite imagery | 82 |