Non‑Myopic Neuro‑Symbolic AI for Smart Infrastructure by Prof Abhishek Dubey

Abstract

Modern cities are large‑scale cyber‑physical systems where sensing, embedded computation, networks, and human operators interact across transportation and energy infrastructure. Current control systems often focus on short‑term goals, leading to unintended long‑term effects. This talk introduces non‑myopic decision‑making, which explicitly accounts for planning uncertainty, disruptions, and long‑horizon impacts. It presents neuro‑symbolic AI methods that combine machine learning with rule‑based reasoning, with applications in vehicle‑to‑grid coordination, microgrid scheduling, transit disruption management, and emergency response. The talk highlights foundational principles and shows how these approaches address complex scalability challenges.

About the Speaker

Abhishek Dubey is an Associate Professor of Computer Science, Electrical Engineering, and Computer Engineering at Vanderbilt University, where he also serves as Associate Dean for Research and Chancellor Faculty Fellow. He directs the Scope Lab, focusing on AI‑driven decision support for resilient cyber‑physical infrastructure. His research spans transportation, energy, and public safety systems, with applications supported by NSF, DARPA, DOE, and NASA. He has also served as Program Director at the U.S. National Science Foundation, overseeing Cyber‑Physical Systems and Smart & Connected Communities programs.