The DSI Smart Cities Center will convene a series of programs focused on the future of urban systems and the data science and AI-driven technologies shaping them. The center supports research that addresses the challenges of aging infrastructure while advancing innovations in smart grids, intelligent transportation systems, and networked sensing technologies that enable real-time monitoring and decision-making.
Through its event programming, the center will bring together researchers exploring how AI, machine learning, and emerging digital infrastructure can improve the resilience, efficiency, and sustainability of dense urban environments. Topics will span predictive maintenance, mobility optimization, energy distribution, urban analytics, and the evolving role of intelligent systems in shaping everyday city life.
The series is open to Columbia University faculty members and affiliated senior researchers. If you’d like to join these meetings, contact Erin Elliott, Events and Marketing Coordinator at ee2548@columbia.edu to receive the location and calendar invite.
Registration Request Form: Registration will be prioritized for Columbia faculty and affiliated scholars. If you are a postdoctoral researcher, student, or external guest, you may be waitlisted to attend until closer to the event date. All who submit on the below form will receive a confirmation email and a calendar hold if your registration is approved.
Speaker and Talk Information Coming Soon
Date of Visit: Wednesday, April 22 (10:00 AM – 11:30 AM)
Speaker: Sam Schwartz, Chair, Sam Schwartz Transportation Research Program, Roosevelt House Public Policy Institute at Hunter College; Founder Sam Schwartz Pedestrian Traffic Management Services; Former New York City Traffic Commissioner
Abstract: In 1970 President Richard Nixon signed the Clean Air Act that, for the first time, regulated mobile emissions i.e. motor vehicles. New York City and scores of other cities were found to be non-attainment zones. The city responded with a plan to toll traffic on all entry points to Manhattan. Lawsuits followed but environmentalists won every battle. The city was ordered by federal courts to implement the tolls by 1978. Nothing could stop the program short of an Act of Congress. Yep, that’s what happened! Further attempts were made over the next 40 years-all failed. Why these plans failed will be discussed as well as why the latest plan succeeded.
But now the city is facing another challenge to managing traffic with the advent of driverless cars, taxis and trucks. Schwartz will discuss the good, bad and ugly scenarios that could unfold and offer recommendation, not just for NYC, but for the country.
Date of Visit: Thursday, March 26 (12:00 PM – 1:00 PM)
Speaker: Minyi Huang, Professor, School of Mathematics and Statistics, Carleton University
Abstract: As an active research area for about two decades, mean field game (MFG) theory provides a powerful tool to tackle large-population non-cooperative dynamic games with individual-mass interactions. This talk describes how to extend MFG theory to models with subpopulations (i.e., agent clusters) distributed over large networks: (i) dense networks, (ii) moderately sparse networks, and (iii) very sparse networks.
The analysis of the dense case (i) has been developed based on graphon theory (Lovasz, 2012), leading to the so-called graphon mean field games (GMFGs). For the second case (ii), by suitable scaling along the graph sequence, we define the limit edge distribution called graphexon, and further introduce a probability kernel to characterize network connections seen at a given node. Then the GMFG theory can be extended to this case. For the very sparse case (iii), we consider a lattice structure to model interactions of neighboring clusters of agents. We apply scaling of diffusion type to obtain a meaningful limit, leading to the so-called Laplexion dynamics, as a hydrodynamic limit of the lattice model. For each case, the limit model will be used to construct decentralized strategies for the actual finite population, which are shown be an asymptotic Nash equilibrium.
(Based on recent work with P. E. Caines, T. Chen, and S. Gao)
Date of Visit: Tuesday, March 10, 2026 (2:00 PM – 3:00 PM)
Speaker: Arturo Deza, Co-Founder and CEO, Artificio
Abstract: As Self-Driving Cars continue to be deployed in different cities around the world: how well will these systems generalize when exposed in new geographies? Moreover, how well will current multi-modal VLMs (Vision Language Models) be able to cognitively understand and act when faces with bizzare edge-case scenarios. In this talk I will aim to answer these questions through a Visual Question Answering (VQA) framework, where we show humans and VLMs a series of our own recorded dashcam footage from Lima and New York City and test for system divergence and convergence. Moreover we tests for these similarities/divergences in a factorial analysis with 3 groups: Humans from NYC, Humans from Lima and VLMs; and two first-person dashcam data recorded from both Lima and New York City.
Date of Visit: Wednesday, February 18, 2026
The Smart Cities Research Center hosted Mr. Lee Shan-Chih, Director-General of the Legal Affairs Bureau of Taichung City Government, Taiwan to foster academic exchange and policy dialogue. The visit focused on Columbia’s work on AI governance, legal frameworks, and public-sector applications of artificial intelligence.
Date of Talk: Tuesday, February 17, 2026
Speaker: Yinhai Wang, Thomas and Marilyn Nielsen Endowed Professor in Engineering, Director of PacTrans and STAR Lab, University of Washington
Abstract: Despite decades of investment and conventional approaches, transportation systems continue to face persistent challenges in safety and mobility. There is an urgent need for more effective,deployment-ready solutions that leverage emerging technologies such as Artificial Intelligence (AI) and the Internet of Things (IoT). However, many existing solutions rely on cloud-centric architectures that suffer from high and unpredictable latency, bandwidth limitations, and substantial deployment and operational costs. These limitations highlight a growing need for affordable, edge-based technologies that can operate reliably on site in real-world transportation environments.
In this talk, the speaker will present his team’s research on edge computing and customized AI methods for mobility and safety applications at the University of Washington’s Smart Transportation Applications and Research Laboratory (STAR Lab). Several representative efforts aimed at transforming roads into safer, smarter, and more community-centered mobility systems will be highlighted, including the award-winning Mobile Unit for Sensing Traffic (MUST) edge AI system, Agentic Edge, and Edge-Retrieval Augmented Generation (Edge-RAG). Strategies for tailoring edge systems to transportation tasks—such as scenario-specific customization and the deployment of lightweight models—will also be discussed.
Date of Visit: Thursday, February 15, 2036 The Smart Cities Research Center hosted a delegation from Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT), along with representatives from the Mitsubishi Research Institute and International Access Corporation, for a research exchange on next-generation urban transportation technologies. The visit highlighted Columbia’s work on hybrid digital twins and the COSMOS testbed, while the delegation shared perspectives on related projects underway in Japan.
Date of Visit: Thursday, February 15, 2036
The Smart Cities Research Center hosted a delegation from Japan’s Ministry of Land, Infrastructure, Transport and Tourism (MLIT), along with representatives from the Mitsubishi Research Institute and International Access Corporation, for a research exchange on next-generation urban transportation technologies. The visit highlighted Columbia’s work on hybrid digital twins and the COSMOS testbed, while the delegation shared perspectives on related projects underway in Japan.
Columbia Presentations:
Date of Talk: Thursday, December 4, 2025
Speaker: George Pechlivanoglou, Director General, Eunice Energy Group; and CEO and President, Joltie
Abstract: Europe’s energy transition has outpaced its digital transition. As renewable penetration rises and grid inertia vanishes, traditional models of forecasting and control collapse under the weight of nonlinear complexity. The European energy market, an intricate, data-rich ecosystem of zonal prices, balancing mechanisms, and infrastructural constraints, demands better data analytics.
This lecture presents a new frontier: the creation of a secure, full-scale digital twin of the European power system, capable of modeling the network itself. By unifying market dynamics with physics-based models of substations, transmission lines, and generation assets, the digital twin reconstructs and simulates the grid’s hidden state in near real time. This model extends beyond energy forecasting toward predictive energy economics and provides a framework to model not only how the grid behaves, but how it will behave under stress.
Modern grids face not only physical instability from renewables but systemic threats from cyberattacks, cascading failures, and coordinated disruptions. Within the digital twin, artificial damage simulations, including cyber-physical intrusion scenarios, enable testing of defense mechanisms, recovery strategies, and autonomous response models. By integrating telemetry, behavioral anomaly detection, and threat intelligence into the energy models, we can prototype techniques with broad applicability across the EU.
Eunice Energy Group provides a unique empirical base for this study by acting as an operational and asset-level data across the Mediterranean’s most dynamic and vulnerable systems. Greece stands as a living laboratory ideal for this type of study: a semi-isolated grid, high in renewable volatility, where resilience and cybersecurity are operational necessities.
The challenge to the data science community is to design the intelligence layer of Europe’s next-generation energy infrastructure through analysis and integration of modalities that have never been combined before with far-reaching individual benefit. Through causal inference, graph neural networks, hybrid AI–physics modeling, and adversarial resilience simulation, data previously restricted to an energy company’s internal frameworks can be unique opportunities for data scientists interested in green energy and climate change.