JULAIN Talk by Arvind Narayanan

Arvind Narayanan
Princeton University, Center for Information Technology Policy
Overcoming pitfalls in the use of machine learning in science

December 9, 2024 | 15:00 CET
VC: https://fz-juelich-de.zoom.us/j/65282324926?pwd=GGiaYJEf5YXo8ZAIJF5vO1Q1MzuXpr.1
Meeting-ID: 652 8232 4926
Kenncode: 955759

Calendar file attached for convenience

Abstract
Most scientific fields are adopting AI rapidly. I argue that this adoption has at times been careless and has led to proliferation of errors such as leakage. These errors lead to flawed models and mistaken scientific findings. To realize the benefits of AI in science and to maintain credibility in the scientific enterprise, a course correction is needed. Scientific fields should adopt AI in a slower and more deliberate fashion so that they can make the necessary changes to scientific epistemology. I present a few other ideas, including checklists that can help catch common pitfalls.

Bio
Arvind Narayanan is a professor of computer science at Princeton University and the director of the Center for Information Technology Policy. He is a co-author of the book AI Snake Oil and a newsletter of the same name which is read by 40,000 researchers, policy makers, journalists, and AI enthusiasts. He previously co-authored two widely used computer science textbooks: Bitcoin and Cryptocurrency Technologies and Fairness in Machine Learning. Narayanan led the Princeton Web Transparency and Accountability Project to uncover how companies collect and use our personal information. His work was among the first to show how machine learning reflects cultural stereotypes, and his doctoral research showed the fundamental limits of de-identification. Narayanan was one of TIME's inaugural list of 100 most influential people in AI. He is a recipient of the Presidential Early Career Award for Scientists and Engineers (PECASE).