|
|
September 3, 2026
|
|
|
Faculty of Computer Science, AGH University of Krakow
|
We invite you to the next seminar in the “Nature-Inspired Computing Methods” series. Our guest will be Roberto Corizzo from American University in Washington, DC, who will give a talk entitled “Open World Anomaly Detection”.
Anomaly detection plays a critical role in many domains where data and system behaviour are constantly evolving, including cyber-physical systems, industrial monitoring, cybersecurity, healthcare, and network traffic analysis. Traditional anomaly detection methods are typically designed to work with static datasets or to continuously adapt to incoming data. Adaptation alone, however, does not guarantee that previously acquired knowledge is preserved.
During the seminar, Roberto Corizzo will introduce the concept of continual anomaly detection, based on continual/lifelong learning. This approach aims to develop machine learning systems that acquire knowledge over time, adapt to new conditions, and retain previously learned information. The talk will cover adaptation strategies, methods for evaluating such systems, and challenges arising from evolving data distributions, including catastrophic forgetting and changes in the underlying notion of “normality”.
Roberto Corizzo is an Assistant Professor at American University in Washington, DC. He received his Ph.D. in Computer Science from the University of Bari Aldo Moro, Italy, in 2018. His research focuses primarily on machine learning and large-scale data analytics, with particular emphasis on methods that enable systems to operate and learn in dynamic and evolving environments.
His research interests include anomaly detection, continual and lifelong learning, time-series and spatio-temporal data analysis, and predictive modelling. The methods he develops have applications in areas including energy, cybersecurity, astrophysics, and social network analysis. He has authored and co-authored over 85 scientific publications and is actively involved in the international research community as an editor and reviewer for journals and conferences in machine learning and data analytics.
Online streaming:
https://agh-mche.webex.com/meet/informatyka
The seminar is free of charge and open to everyone interested – including staff, PhD students, students, and participants from outside AGH.
We warmly invite you to join us!