Adversarial Robustness Toolbox (ART)
Date:
Presentation: HDR Research Tool Symposium Presentation
Tools make our research easier and faster, and research tools range from ChatGPT to the latest neural network architectures to advanced simulation software. Hear about the tools that EECS Higher Degree by Research (HDR) students are using in their research (applications, advantages and limitations) and how they differentiate between their contributions to their field and the capabilities provided by these tools. The symposium will start with 5 minute lightning talks from EECS HDRs, then follow with a stimulating panel discussion with experts in their field.
At this symposium, I gave a talk about the Adversarial Robustness Toolbox (ART), a Python Library for Machine Learning Security, providing tools that enable developers and researchers to defend and evaluate Machine Learning models and applications against the adversarial threats of Evasion, Poisoning, Extraction, and Inference. All while being model, implementation, tasks and data type agnostic!