IEM Distinguished Lecturers Seminar: An AI Foundation Model Approach to Signal Processing and Control in Cyber-Physical Systems
***Welcome Coffee & Sweets at 13:00***
Abstract
AI Foundation models are learning, reasoning, and decision-making architectures trained on large datasets and adaptable across diverse applications. Most successful AI foundation models are large language models that are pre-trained on text, image, and video data to capture linguistic and contextual information. Applying such models to cyber-physical systems—governed by physical rather than linguistic laws—presents unique challenges, especially in causal signal processing, control, and decision-making tasks. This talk explores an alternative physics-based AI foundation model inspired by the Wiener-Kallianpur innovation representation of stationary processes. Pre-trained on time series generated by physical systems, the proposed foundation model extracts innovations for adaptions across multiple tasks, including probabilistic forecasting, anomaly detection, and system protection.
Short Bio
Lang Tong is the Irwin and Joan Jacob Professor of Engineering at Cornell University and the Site Director of the Power System Engineering Research Center. His current research focuses on power system optimizations, electricity markets, and AI/machine learning technologies for power system operations. He received a B.E. in Automation from Tsinghua University, a Ph.D. from the University of Notre Dame, and was a Postdoctoral Research Associate at Stanford University. A Fellow of IEEE, he was the 2018 Fulbright Distinguished Chair in Alternative Energy.
Abstract
AI Foundation models are learning, reasoning, and decision-making architectures trained on large datasets and adaptable across diverse applications. Most successful AI foundation models are large language models that are pre-trained on text, image, and video data to capture linguistic and contextual information. Applying such models to cyber-physical systems—governed by physical rather than linguistic laws—presents unique challenges, especially in causal signal processing, control, and decision-making tasks. This talk explores an alternative physics-based AI foundation model inspired by the Wiener-Kallianpur innovation representation of stationary processes. Pre-trained on time series generated by physical systems, the proposed foundation model extracts innovations for adaptions across multiple tasks, including probabilistic forecasting, anomaly detection, and system protection.
Short Bio
Lang Tong is the Irwin and Joan Jacob Professor of Engineering at Cornell University and the Site Director of the Power System Engineering Research Center. His current research focuses on power system optimizations, electricity markets, and AI/machine learning technologies for power system operations. He received a B.E. in Automation from Tsinghua University, a Ph.D. from the University of Notre Dame, and was a Postdoctoral Research Associate at Stanford University. A Fellow of IEEE, he was the 2018 Fulbright Distinguished Chair in Alternative Energy.
Practical information
- General public
- Free