Privacy Testing for Deep Learning

A micro-service reference test application for model extraction, cloud management, energy efficiency, power prediction, single- and multi-tier auto-scaling

General-purpose library for extracting interpretable models from Multi-Agent Reinforcement Learning systems

Analytic tableau based minimal model generator, model checker and theorem prover for first-order logic with modal extensions

MEME: Generating RNN Model Explanations via Model Extraction

Simple machine learning in Python/Tensorflow with model saving

For our AAAI23 paper "DisGUIDE: Disagreement-Guided Data-Free Model Extraction" (Oral Presentation) by Jonathan Rosenthal, Eric Enouen, Hung Viet Pham, and Lin Tan.

Ecore metamodel reverse engineering: Automatically extract EMF metamodels from Java code.

📄 [Talk] OFFZONE 2022 / ODS Data Halloween 2022: Black-box attacks on ML models + with use of open-source tools

CME: Concept-based Model Extraction

Marich is a model-agnostic extraction algorithm. It uses a public data to query a private model, aggregates the predicted labels, and construct a distributionall equivalent/max-information leaking extracted model.

A neural network model builder, leveraging a neuro-symbolic interface.

The Labelled Transition Systems Extractor tool project

Minimal reproducible PoC of 3 ML attacks (adversarial, extraction, membership inference) on a credit scoring model. Includes pipeline, visualizations, and defenses

Model Reconstruction from Counterfactual Explanations

Graphical User Interface to debug ROS systems

Comprehensive model extraction attack

Code for VidModEx: Interpretable and Efficient Black Box Model Extraction for High-Dimensional Spaces

Serverless Application Extraction System (SAES)

Collection of the TeX files and figures used to create my UofA CS master's thesis