On Friday, November 13, a seminar will be held in Fassò Room (Campus Leonardo, Building 4A, floor 1), starting at 16:30 CET.

The seminar will be given by Dr. Thilanka Munasinghe, Rensselaer Polytechnic Institute, and is titled “Bridging Quantum Machine Learning and Spatial Systems: Harnessing Quantum Computing for Geographic and Spatial Applications”.

Abstract
Quantum technologies are rapidly advancing, with quantum computing, quantum communication, and quantum optimization set to revolutionize next-generation communication networks and learning systems. As quantum computing evolves from theoretical ideas to hardware prototypes capable of managing hundreds of qubits, recent advancements in industry and academia have enabled promising applications in classification, regression, generative modeling, and reinforcement learning. Quantum Machine Learning (QML) stands as an innovative field at the intersection of Quantum Computing (QC) and Artificial Intelligence (AI). This presentation explores the transition from classical machine learning to quantum-driven spatial systems within the current Noisy Intermediate-Scale Quantum (NISQ) era. We will examine how Variational Quantum Algorithms (VQAs) and Quantum Kernel methods can address high-dimensional environmental data for NISQ devices (e.g., IBM quantum systems) and explore their associated challenges. We will examine data inputs from NASA missions such as the precipitation data from Global Precipitation Measurement (GPM) mission, Land Surface Temperature (LST), and Soil Moisture observations data in QML applications, alongside current benchmarking with classical ML models. The talk addresses strategies for managing quantum hardware constraints and implementing QML models for classification and regression tasks related to floods, wildfires, and other domains such as agricultural crop-frosting. Attendees will gain a functional understanding of how to transform classical datasets into quantum states and choose appropriate quantum encoding methods for spatial data analysis. The talk will provide a foundation for understanding how to choose QML models to run on both quantum simulators and real quantum computing hardware for geospatial applications.

Speaker’s bio
Dr. Thilanka Munasinghe is a faculty member and Lead Research Specialist at Rensselaer Polytechnic Institute (RPI). He holds a BSc in Aerospace Engineering and an MSc in Mechanical Engineering from West Virginia University, and a PhD in Information Science from the University at Albany, New York. Currently, his research interests are in developing QML applications that leverage the intersection of Quantum Computing (QC) and Artificial Intelligence (AI), harnessing quantum algorithms. He serves on the organizing committee for international conferences and workshops on Quantum & AI for industry applications and scientific discovery.

The seminar can also be followed remotely at the following link: https://politecnicomilano.webex.com/meet/alberto.vavassori