CEMSE Weekly Updates - May 5, 2026 Tue, May 5 2026 Newsletter Upcoming Events Stay informed about the upcoming events and the latest news from CEMSE. Learning under Limited Information across Federated, Multi-Agent, and LLM Settings Salma Kharrat, Ph.D. Student, Computer Science May 7, 15:00 - 16:45 B3 R5220 Federated learning personalized learning decentralized learning Reinforcement Learning black-box optimization prompt optimization decentralized systems combinatorial optimization observability inference Trustworthy AI trustworthy machine learning intelligent systems LLM This dissertation studies learning under structural information constraints across three major paradigms: federated learning, cooperative multi-agent reinforcement learning, and black-box optimization of large language models. Intelligent Multi-Patient Cardiac Monitoring System with Real-Time Photonic Edge Processing and Wearable Multimodal Sensing Zhican Zhou, Ph.D. Student, Electrical and Computer Engineering May 10, 12:00 - 13:00 B9 R2325 Stability and Signal Generator Agnostic Moment Matching Alessandro Astolfi, Professor, Applied Mathematics and Computational Science May 10, 14:30 - 15:30 B1 R3119 moment matching model reduction interpolation Stability This talk explores advancements in model reduction that modernize traditional moment matching by introducing a data-driven procedure for unknown signal generators and a closed-loop interpolation framework that extends these techniques to unstable systems. Differentiable Optics for Automated Optical Design and End-to-end Computational Imaging Xinge Yang, Ph.D. Student, Computer Science May 11, 17:00 - 18:30 B1 R2202 differentiable optics computational imaging optical design automated design augmented reality virtual reality AR VR This thesis develops a differentiable optics and image simulation framework, with applications in automated lenses and AR/VR design and end-to-end computational imaging with novel camera systems. Enabling Seamless Connectivity via Virtualization Technologies in 6G Integrated Networks Sahar Ammar, Ph.D. Student, Electrical and Computer Engineering May 13, 12:00 - 14:00 B1 R4214 Reinforcement Learning networks Non-Terrestrial Networks Cellular Networks Virtualized Network Function 6g wireless systems machine learning network slicing This thesis investigates how integrated networks powered by such technologies can be designed and optimized to enable seamless connectivity. Time Series Clustering: Pattern Recognition, Forecasting, and Amortized Inference Ángel López Oriona, Postdoctoral Research Fellow, Statistics May 14, 12:00 - 13:00 B9 R2325 Time Series Pattern Recognition forecasting statistical inference This talk presents innovative time series clustering techniques, highlighting a quantile-based approach for analyzing locally stationary data, a predictive framework for enhanced forecasting, and the use of amortized inference to overcome traditional algorithmic limitations. Bayesian Spatio-Temporal Modeling for Environmental Monitoring and Epidemiology: Disaggregation and Disease Spread Dynamics Fernando Rodriguez Avellaneda, Ph.D. Student, Statistics May 14, 14:00 - 16:00 B3 R5220 This PhD thesis introduces innovative statistical frameworks for modeling and interpreting spatial and spatio-temporal dynamics in geostatistical data and point processes, with applications in air pollution monitoring and infectious disease dynamics. The first project focuses on spatial disaggregation of normally distributed multivariate geostatistical data, motivated by air pollution applications in Portugal and Italy. The second project extends spatial disaggregation to a spatio-temporal setting for univariate normally distributed data, with the methodology illustrated through the monitoring Scalable Methods for Multivariate Normal Probability Estimation with Applications in Confidence Region Detection, Transport Phenomena, and Parallel Computing Using RCOMPSs Xiran Zhang, Ph.D. Student, Statistics May 14, 15:00 - 17:00 B2 R5209 geospatial statistics spatio-temporal statistics High Performance Computing HPC multivariate statistics GPU Algorithms This thesis addresses computing high-dimensional multivariate normal (MVN) probabilities in environmental and geospatial applications by combining high-performance computing, numerical approximation, and transport-based covariance modeling to make several important spatial procedures usable at larger scales.
