Title: Domain-Constrained Inference from Sensor Data in Civil Infrastructure Systems
In some sense we are living in an incredibly exciting time to do research and really understand and manipulate the world around us (though most people in the past would say the same). Our physical environment, especially the built environment, is being covered with sensors, actuators and computing devices at an incredibly fast rate. Likewise, we have developed highly-flexible and efficient mathematical models and tools to make sense of all the data that these devices are generating. This marriage of high-volumes of data with high-capacity and efficient models for inference have resulted in groundbreaking advances in areas such as computer vision and natural language processing. However, the same marriage has not had similar results in other domains with lower data availability and/or where physics imposes more difficult constraints on the inverse problem, such as in most applications in the civil infrastructure systems domain. In this talk I discuss how we can incorporate domain-expertise to improve our ability to perform (approximate) inference in these settings, and will illustrate these ideas through two case studies: one on the problem of disaggregating electrical power measurements in a home, and another on the task of detecting structural defects on bridges via vibration measurements collected on vehicles moving across the bridge.
About the speaker
Mario Bergés is an associate professor in the Department of Civil and Environmental Engineering at Carnegie Mellon University (CMU). He is interested in making our built environment more operationally efficient and robust through the use of information and communication technologies, so that it can better deal with future resource constraints and a changing environment. He has co-authored more than 20 peer-reviewed journal publications and over 80 conference/workshop articles on topics related to sensing and data analysis for civil infrastructure systems, with a particular focus on buildings and energy efficiency. Prof. Bergés is the faculty co-director of the Smart Infrastructure Institute at CMU, as well as the director of the Intelligent Infrastructure Research Lab (INFERLab). Among recent awards, he received the Professor of the Year Award by the ASCE Pittsburgh Chapter in 2018, Outstanding Early Career Researcher award from FIATECH in 2010, and the Dean's Early Career Fellowship from CMU in 2015. He received his B.Sc. in 2004 from the Instituto Tecnológico de Santo Domingo, in the Dominican Republic; and his M.Sc. and Ph.D. in Civil and Environmental Engineering in 2007 and 2010, respectively, both from Carnegie Mellon University.