From: 2025-06-04 14:15 to 15:00
Place: Seminar Room M 3170-73 in the M-building, LTH
Contact: emma [dot] tegling [at] control [dot] lth [dot] se

Date & Time: June 4th, 14:15-15:00
Location: Seminar Room M 3170-73 at Dept. of Automatic Control, LTH
Speaker: Giacomo Baggio
Title: Input-Driven Memory Retrieval in Hopfield Networks
Abstract:  Introduced in the 1980s, the Hopfield model offers a biologically plausible framework for modeling associative memory and has since inspired extensive research in neuroscience and machine learning. In this talk, I will begin by reviewing the classical Hopfield model, with a focus on its dynamical properties. Building on this, I will explore how external inputs shape network dynamics and introduce a novel input-driven plasticity mechanism aimed at improving memory retrieval. I will analyze the dynamical behavior of the resulting Input-Driven Plasticity (IDP) Hopfield model and illustrate its performance in terms of retrieval accuracy and robustness to noise. The talk is based on joint work with Simone Betteti, Francesco Bullo, and Sandro Zampieri.

Bio: Giacomo Baggio is an Associate Professor in the Department of Information Engineering at the University of Padova, Italy. He received his Ph.D. in Information Engineering from the University of Padova in 2018. He has held research positions as a Visiting Scholar at the University of Cambridge, UK, and as a Postdoctoral Scholar at the University of California, Riverside, USA. His research interests are in the area of systems and control theory, with applications to network science and data-driven modeling. His work has been recognized with awards including the Roberto Tempo Best CDC Paper Award and the IEEE Control Systems Letters Outstanding Paper Award.