Sourav Medya

Assistant Professor, Department of Computer Science, University of Illinois Chicago

prof_pic.jpg

Email: medya at uic dot edu

My research focuses on problems at the intersection of graphs, machine learning, and data science, with an emphasis on interpretability, effciency, and safety. I also build machine learning techniques with applications in healthcare, infrastructure, and computational social science.

Before joining UIC, I was a research assistant professor in the Kellogg School of Management at Northwestern University and the Northwestern Institute on Complex Systems. I received my Ph.D. in Computer Science from the University of California, Santa Barbara, and my Master of Engineering degree from the Department of Computer Science and Automation at the Indian Institute of Science, Bangalore.

For the full publication list, see Research.

selected publications

  1. TMLR
    Freeze, Prompt, and Adapt: A Framework for Source-free Unsupervised GNN Prompting
    Peyman Baghershahi and Sourav Medya
    Transactions on Machine Learning Research, 2026
    Link
  2. ACL
    From Nodes to Narratives: Explaining Graph Neural Networks with LLMs and Graph Context
    Peyman Baghershahi, Gregoire Fournier, Pranav Nyati, and Sourav Medya
    In The Association for Computational Linguistics, 2026
    Link
  3. KDD
    RIDGECUT: Learning Graph Partitioning with Rings and Wedges
    Qize Jiang, Angelo Zangari, Linsey Pang, Alice Gatti, Mahima Aggarwal, Giovanna Vantini, Xiaosong Ma, Weiwei Sun, Sourav Medya, and Sanjay Chawla
    In ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2026
    Link
  4. ICLR
    Learning Exposure Mapping Functions for Inferring Heterogeneous Peer Effects
    Shishir Adhikari, Sourav Medya, and Elena Zheleva
    In International Conference on Learning Representations, 2026
    Link
  5. ICML
    COMRECGC: Global Graph Counterfactual Explainer through Common Recourse
    Gregoire Fournier and Sourav Medya
    In International Conference on Machine Learning, 2025
    Link