Devashish Khulbe

Devashish Khulbe
Email CV Scholar

I am a researcher specializing in Deep Learning and its applications. My current research focuses on developing efficient, high-precision graph representations using advanced machine learning models. These are applied in addressing real-world challenges in domains such as transportation and social sciences, aiming to enhance predictive accuracy and operational efficiency.

I have collaborated closely with researchers at New York University, Masaryk University, and Thales Group and have been involved in several interdisciplinary projects that bridge AI and data science with real-world applications. My work has been focused on developing novel approaches to graph neural networks, with particular emphasis on interpretability and applications in urban systems.

Some key areas of my research include:

  • Graph Representation Learning: Developing efficient architectures for large-scale graph data representation and modeling, including knowledge graphs and ontologies.

  • Urban Informatics: Using AI to understand complex social dynamics and behavioral patterns in Urban Systems.

Publications

* indicates equal contribution.

Urban delineation through the lens of commute networks: Leveraging graph embeddings to distinguish socioeconomic groups in cities
Devashish Khulbe, Stanislav Sobolevsky
PLOS Complex Systems, 2025
[ link ]
Commute Networks as a Signature of Urban Socioeconomic Performance: Evaluating Mobility Structures with Deep Learning Models
Devashish Khulbe, Alexander Belyi, Stanislav Sobolevsky
Smart Cities, 2025
[ link ]
Distance deterrence comparison in urban commute among different socioeconomic groups: A normalized linear piece-wise gravity model
Mingyi He, Yuri Bogomolov, Devashish Khulbe, Stanislav Sobolevsky
Journal of Transport Geography, 2023
[ link ]
Mobility Networks as a Predictor of Socioeconomic Status in Urban Systems
Devashish Khulbe, Alexander Belyi, Ondřej Mikeš, Stanislav Sobolevsky
International Conference on Computational Science and Its Applications, 2023
[ link ]
A probabilistic simulation framework to assess the impacts of ridesharing and congestion charging in New York City
Devashish Khulbe*, Chaogui Kang*, Satish Ukkusuri, Stanislav Sobolevsky
Data Science for Transportation, 2023
[ link ]
Impact of income on urban commute across major cities in US
Yuri Bogomolov, Mingyi He, Devashish Khulbe, Stanislav Sobolevsky
Procedia Computer Science, 2023
[ link ]
Modeling Severe Traffic Accidents with Spatial and Temporal Features
Soumya Sourav, Devashish Khulbe, Vishal Verma
International Conference on Neural Information Processing, 2019
[ link ]

Teaching

Spring 2021, 2022 (NYU CUSP):
Applied Data Science
(Adjunct Instructor)
Spring 2020, 2021, 2022 (NYU CUSP):
Principles of Urban Informatics
(Adjunct Instructor)

News

(Jul 25): Our paper on Urban socioeconomic modeling using GNNs was accepted at Smart Cities.
(Jul 25): Our paper on Urban Delineation using Graph Representation Learning was accepted at PLOS Complex Systems.
(Jan 25): I joined as a researcher at Thales Research and Technology, working on network AI models for knowledge graphs.
(Dec 24): I presented our research on modeling mobility networks with deep learning methods at Complex Networks 2024 conference in Istanbul, Turkey.
(Sep 24): I am part of the core team starting the Data Analytics undergraduate program at MUNI, beginning Fall 24.
(Jul 23): I presented our paper Mobility Networks as a Predictor of Socioeconomic Status in Urban Systems at ICCSA 2023 in Athens, Greece.
(May 23): I presented our research with commute network modeling with Graph Neural Networks in FRCCS 2023 in Le Havre, France.