by Trung Nguyen | Mar 1, 2024 | Materials, News, Research
Under the supervision of Professor Gill Dobbie and Dr Vithya Yogarajan, this research project by Kejun Dai reviews frameworks for evaluating bias in artificial intelligence models. Situated within the broader “Ethical Computing” initiative led by Professor...
by Trung Nguyen | Dec 19, 2023 | Materials, News
Image: A painting of a futuristic nurse in rural New Zealand created using Stable Diffusion on huggingface.co This recent report on AI and healthcare is from the Office of the Prime Minister’s Chief Science Advisor. FOREWORD: This report was requested by Prime...
by Trung Nguyen | Nov 29, 2023 | Materials, News, Research
Gill delivered a presentation at the ARC Training Centre in Data Analytics for Resources and Environments (DARE), The University of Sydney, where she shared insights from her research focused on quantifying bias in language models with the ultimate goal of debiasing....
by Trung Nguyen | Nov 22, 2023 | Materials, News
Following our recent post, the group of “Teaching materials for AI Ethics” has successfully presented at the Capstone event (photo above). Project description: This year, the team EthicAI developed a resource with real-world use cases of the latest AI...
by Trung Nguyen | Nov 10, 2023 | Materials, News
This semester, two groups of students from the School of Computer Science Capstone course (COMPSCI 399) worked on a project aiming at developing teaching materials for AI Ethics. The goal was to develop an interactive website that introduces students to the potential...
by Trung Nguyen | Oct 25, 2023 | Materials, News
(Figure courtesy of Dr Ali Hatamizadeh, the author of GradViT) Recent studies have shown that training samples can be recovered from the corresponding gradients. This phenomenon is called Gradient Inversion (GradInv) attacks. The principle behind GradInv is to learn...