Presenter: Pedro Casas
Friday, February 28th, 11:00 AM
Location: Sala Grande Covivio, Corso Ferrucci 112
ABSTRACT
AI4Industry is transforming the European industrial landscape, driving efficiency, innovation, and competitive advantage across sectors such as energy, automotive, aerospace, and manufacturing. By leveraging AI/ML and data-driven analysis, industries can enhance productivity, optimize operations, reduce carbon footprint, and unlock new customer value while reducing costs. The rapid adoption of AI in industry enables complex-process modeling, real-time analytics, predictive maintenance, anomaly detection, and in general intelligent decision-making in complex, data-intensive environments.
However, challenges hindering both practitioners and academic research in this domain remain; AI4Industry must contend with complex, heterogeneous data sources and stringent reliability requirements. Industrial data is often fragmented, context-dependent, and requires problem-specific datasets that are typically difficult to acquire. AI/ML solutions in this domain tend to be highly tailored, making the end-to-end development and applicability process tedious, time-consuming, and difficult to generalize.
In this talk, Pedro will first illustrate the potential of AI to optimize industrial operations through a general presentation of real-world AI4Industry applications they are currently addressing in AIT. He will then focus on a particular type of data and problems which are common to many industries, namely the analysis of time-series data. Traditional approaches to time-series analysis often struggle to manage the scale, complexity, and dynamism of modern time-series data problems. He will therefore explore how they are using deep learning and foundation models to tackle time-series analysis in the industry, in particular to detect anomalies and classify time-series in real-time monitoring applications.
BIOGRAPHY
Dr. Pedro Casas is Senior Scientist at the AIT Austrian Institute of Technology, Vienna, leading the AI4NETS and the AI4Industry Research Areas – AI/ML for Networking and Industrial Applications. He received an Electrical Engineering degree from Universidad de la República, Uruguay, in 2005, a Ph.D. degree in Computer Science from Télécom Bretagne in 2010, and a Ph.D. degree in Electrical Engineering from Universidad de la República in 2013. He was a Postdoctoral Researcher at LAAS-CNRS, Toulouse, from 2010 to 2011, and a Senior Researcher at the FTW Telecommunications Research Center Vienna from 2011 to 2015.
His work focuses on AI/ML-based approaches for networking applications, big data analytics and platforms, Internet network measurements, network security and anomaly detection, as well as Internet QoE monitoring. He has published more than 220 networking research papers in major international conferences and journals, and received 18 awards for his work, including eight Best Paper Awards. He has been technical work leader and contributor in 32 different national and international projects, including both research and commercial projects. Besides managing industry-funded projects, he is currently acting as PI of the EU GRAPHS4SEC project and the Austrian FFG AI4SimProd project. He drives the organization of different actions in network measurement and analysis, including the IEEE ComSoc ITC Special Interest Group on Network Measurements and Analytics, the IEEE/IFIP TMA Conference, the IEEE S&P WTMC Workshop, and the ACM CoNEXT GNNet Workshop.