Subhankar Ghosh

iHARP Research Assistant | UMN Ph.D. Candidate

Email: ghosh117@umn.edu
See me at Google Scholar Citation Page, LinkedIn

Research Interests

  • Generative AI
  • Multimodal Fusion
  • Anomaly Detection
  • Computer Vision
  • Spatial Statistics
  • Data Mining.
Short Biography

Subhankar Ghosh is a Computer Science Ph.D. candidate at the University of Minnesota working with Professor Shashi Shekhar. His previous projects focused on detecting colocation patterns of entities, such as retail establishments, with statistical guarantees. He is currently collaborating with domain scientists in iHARP on various generative AI approaches to enhance sea-level forecasting.

Publications
  • Mingzhou Yang, Bharat Jayaprakash, Subhankar Ghosh, Hyeonjung Tari Jung, Matthew Eagon, William F. Northrop, and Shashi Shekhar. 2025. Climate smart computing: A perspective. Pervasive and Mobile Computing 108, (March 2025), 102019. https://doi.org/10.1016/j.pmcj.2025.102019
  • Subhankar Ghosh, Arun Sharma, Jayant Gupta, Aneesh Subramanian, and Shashi Shekhar. 2024. Towards Kriging-informed Conditional Diffusion for Regional Sea-Level Data Downscaling: A Summary of Results. In Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems, October 29, 2024. ACM, Atlanta GA USA, 372–383. https://doi.org/10.1145/3678717.3691304
  • Subhankar Ghosh, Arun Sharma, Jayant Gupta, and Shashi Shekhar. 2024. Towards Statistically Significant Taxonomy Aware Co-Location Pattern Detection (Short Paper). LIPIcs, Volume 315, COSIT 2024 315, (2024), 25:1-25:11. https://doi.org/10.4230/LIPICS.COSIT.2024.25 *
  • Subhankar Ghosh, Jayant Gupta, Arun Sharma, Shuai An, and Shashi Shekhar. 2023. Reducing False Discoveries in Statistically-Significant Regional-Colocation Mining: A Summary of Results. In 12th International Conference on Geographic Information Science (GIScience 2023) (Leibniz International Proceedings in Informatics (LIPIcs)), 2023. Schloss Dagstuhl – Leibniz-Zentrum für Informatik, Dagstuhl, Germany, 3:1-3:18. https://doi.org/10.4230/LIPIcs.GIScience.2023.3 [Conference Paper]
  • Subhankar Ghosh, Shuai An, Arun Sharma, Jayant Gupta, Shashi Shekhar, and Aneesh Subramanian. 2023. Reducing Uncertainty in Sea-level Rise Prediction: A Spatial-Variability-Aware Approach. In I-GUIDE Forum 2023: Harnessing the Geospatial Data Revolution for Sustainability Solutions, 2023. Purdue University Library Publishing. https://doi.org/10.5703/1288284317665
  • Subhankar Ghosh, Jayant Gupta, Arun Sharma, Shuai An, and Shashi Shekhar. 2023. Reducing False Discoveries in Statistically-Significant Regional-Colocation Mining: A Summary of Results. In 12th International Conference on Geographic Information Science (GIScience 2023) (Leibniz International Proceedings in Informatics (LIPIcs)), Schloss Dagstuhl – Leibniz-Zentrum für Informatik, Dagstuhl, Germany, 3:1-3:18. DOI: https://doi.org/10.4230/LIPIcs.GIScience.2023.3 [Conference Paper]
  • Subhankar Ghosh, Jayant Gupta, Arun Sharma, Shuai An, and Shashi Shekhar. 2022. Towards geographically robust statistically significant regional colocation pattern detection. In Proceedings of the 5th ACM SIGSPATIAL International Workshop on GeoSpatial Simulation, ACM, Seattle Washington, 11–20. DOI:https://doi.org/10.1145/3557989.3566158 [Conference/Workshop Paper]
Internships/ Fellowships
  • 2025 Spring- Oak Ridge National Lab, Tennessee
    • Subhankar worked on Vision Transformers and Diffusion Models for climate downscaling tasks as part of ORBIT: Oak Ridge Base Foundation Model for Earth System Predictability.
  • 2025 Summer – Amazon, Bellevue (WA)
    • Subhankar will work with generative AI and multimodal fusion for anomaly detection in geospatial applications as an Applied Scientist intern in the Amazon Geospatial team.