The Urban AI lab develops fundamental theory of urban science using artificial intelligence. The lab investigates three research themes in intelligent individual decisions, spatiotemporal urban dynamics, and computational urban justice. The first theme focuses on the individual decisions by integrating discrete choice models and deep learning with wide urban applications in the choice of travel modes, travel routes, residential locations, and urban activities. The second theme treats cities as interrelated system. By integrating network theory and deep learning, it quantifies the spatiotemporal dynamics between people and places, thus facilitating the design of urban systems to advance resilience and sustainability. The third research theme focuses on the normative aspect of urban science by enhancing transparency, accountability, and fairness of the urban machine intelligence to achieve broad social impacts. With the theoretical innovations and practical impacts, the lab seeks to create a more sustainable, intelligent, and equitable urban future with artificial intelligence.
The lab’s work has been published in a wide range of journals and conferences. Our methodological work about integrating discrete choice models and deep neural networks has been published in Transportation Research Part B: methodological, Transportation Research Part C: emerging technologies, and Journal of Choice Modeling. Our work about spatiotemporal urban dynamics has been published in IEEE Transactions on Intelligent Transportation Systems and ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Our work about computational fairness and smart governance has been published in Transportation Research Part A: Policy and Practice and International Journal of Sustainable Transportation. The research has been funded by the US Department of Energy (DOE) and Singapore-MIT Alliance for Research and Technology (SMART).
Due to its highly interdisciplinary nature, the Urban AI Lab contributes to multiple intellectual communities. It contributes to the modeling of individual travel behavior in the fields of urban planning, civil engineering, and decision science. It also connects the behavioral analysis to policy discussions, contributing to the fields of economics, governance, and public policy. Methodologically, the lab also publishes in the leading computer science conferences, thus contributing to the fields of artificial intelligence, machine learning, and data mining. The Urban AI lab has deep collaborations with laboratories around the world, such as MIT Urban Mobility Lab, MIT Media Lab, and Berkeley Institute of Transportation Studies.