AI@JSI Seminar - Multi-Objective AutoML: Towards Accurate and Robust Neural Networks

by Jan N. van Rijn

Tuesday 10 Dec 2024, 13:00 → 14:00 Europe/Ljubljana

A/2-A214 - Velika predavalnica (Jamova)

Dear colleagues,

we kindly invite you to the next AI@JSI seminar, which will take place on Tuesday, 10.12. at 13:00. The seminar will be held in JSI main lecture hall (Jamova 39). You are cordially invited to attend it in person, but remote attendance will also be possible via zoom (https://zoom.us/j/91879263877?pwd=9sxadIBnOrSxnyZdipwLiNNurcta3O.1). The details on the lecture and lecturer are given below. We are very much looking forward to meeting you at the seminar.

-----

Title: Multi-Objective AutoML: Towards Accurate and Robust Neural Networks

Abstract: Automated machine learning has been successful in supporting data scientists in selecting appropriate machine learning architectures, as well as optimizing hyperparameters. By doing so, data scientists can focus their attention on more important tasks. Partially thanks to the TAILOR project, in which Leiden University and JSI have successfully collaborated, we have seen a demand on AutoML techniques to not only provide solutions that are accurate, but also those that are trustworthy according to several relevant criteria. In particular neural networks are known to be vulnerable to adversarial attacks, whereas robustness (against such attacks) is an important criterion of trustworthiness. In this talk, I will summarize various projects we have done through this collaboration, that envision AutoML solutions that specifically address robustness of neural networks.

Lecturer: Jan N. van Rijn holds a tenured position as assistant professor at Leiden University, where he works in the computer science department (LIACS) and Automated Design of Algorithms cluster (ADA). His research interests include trustworthy artificial intelligence, automated machine learning (AutoML) and metalearning.

Google Calendar event: https://bit.ly/41ENEEb

-----

Information on AI@JSI seminars and previous recordings are available here: https://kt.ijs.si/aijsi-seminar/.

With best wishes,

Sašo & Nina