CV

Athresh Karanam

Education

  • May 2026

    Dallas, TX

    Ph.D. in Computer Science
    University of Texas at Dallas
    Dissertation: Effective, Efficient, and Explainable Learning in Data-Scarce and Noisy Domains (PDF)
    Advisor: Prof. Sriraam Natarajan
  • May 2015

    Guwahati, India

    B.Tech. in Civil Engineering
    Indian Institute of Technology, Guwahati
    Advisor: Prof. Subashisa Dutta

Research Interests

  • Deep Tractable Probabilistic Models, Human-Allied Learning, Explainable AI, Data- and Compute-Efficient Learning, Submodular Optimization, Causal Inference, Generalizable ML, AI for Healthcare.

Technical Skills

  • Python, PyTorch, Java, MATLAB, R, Git, SQL.

Conference Papers

  • 2026
    On the Transferability of Agricultural Weed Detection Under Cross-Field Distribution Shift
    International Conference on Dynamic Data Driven Applications Systems (DDDAS 2026)
    Prabhakar, N., Tenali, P., Abudeye Fernandez, W., Borah, S., Karanam, A., Blasch, E., Sundaravadivel, P., Natarajan, S.
  • 2026
    Neurosymbolic Imitation Learning with Human Guidance: A Privileged Information Approach
    International Joint Conference on Learning & Reasoning (IJCLR 2026)
    Prabhakar, N., Balaji, V., Karanam, A., Kersting, K., Natarajan, S.
  • 2025
    A Unified Framework for Human-Allied Learning of Probabilistic Circuits
    Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2025)
    Karanam, A., Mathur, S., Sidheekh, S., Natarajan, S.
  • 2023
    Exploiting Domain Knowledge as Causal Independencies in Modeling Gestational Diabetes
    Pacific Symposium on Biocomputing (PSB 2023)
    Mathur, S., Karanam, A., Radivojac, P., Haas, D. M., Kersting, K., Natarajan, S.
  • 2022
    Orient: Submodular Mutual Information Measures for Data Subset Selection under Distribution Shift
    Advances in Neural Information Processing Systems (NeurIPS 2022)
    Karanam, A., Killamsetty, K., Kokel, H., Iyer, R.
  • 2022
    Explaining Deep Tractable Probabilistic Models: The Sum-Product Network Case
    International Conference on Probabilistic Graphical Models (PGM 2022)
    Karanam, A., Mathur, S., Radivojac, P., Kersting, K., Natarajan, S.
  • 2021
    Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Models
    Advances in Neural Information Processing Systems (NeurIPS 2021)
    Zečević, M., Dhami, D. S., Karanam, A., Natarajan, S., Kersting, K.
  • 2021
    A Probabilistic Approach to Extract Qualitative Knowledge for Early Prediction of Gestational Diabetes
    Artificial Intelligence in Medicine (AIME 2021)
    Karanam, A., Hayes, A. L., Kokel, H., Haas, D. M., Radivojac, P., Natarajan, S.

Workshop Papers

  • 2023
    Test-time Active Feature Selection through Tractable Acquisition Functions
    Tractable Probabilistic Models Workshop at UAI (UAI TPM 2023)
    Karanam, A., Natarajan, S.
  • 2023
    Bayesian Learning of Probabilistic Circuits with Domain Constraints
    Tractable Probabilistic Models Workshop at UAI (UAI TPM 2023)
    Karanam, A., Mathur, S., Sidheekh, S., Natarajan, S.

Tutorials

  • 2024
    Deep Tractable Probabilistic Models
    Tutorial at CODS-COMAD
    Mathur, S., Sidheekh, S., Karanam, A., Natarajan, S.

Research Experience

  • 2020-2026
    Ph.D. Researcher
    Starling Lab, University of Texas at Dallas
    Advisor: Prof. Sriraam Natarajan
    Dissertation research on deep tractable probabilistic models, human-allied learning, and explainability in data-scarce domains. Developed a unified framework for incorporating expert domain knowledge into probabilistic circuits (AAAI 2025); introduced CSI-tree representations for extracting human-interpretable context-specific independencies from learned models (PGM 2022); proposed Orient, a submodular mutual information-based data subset selection framework for domain adaptation with 2-3× training speedups (NeurIPS 2022); and applied probabilistic models to clinical datasets for gestational diabetes prediction and knowledge extraction.
  • 2025
    AI Research Intern
    Altir LLC
    Developed an interpretable, uncertainty-aware LLM agent by augmenting large language models with local tractable probabilistic reasoning, improving reliability through faithful uncertainty quantification.
  • 2020
    Machine Learning Intern
    RelationalAI Inc.
    Built an interactive product recommendation tool using tractable probabilistic models; trained and evaluated boosted relational models on an online vendor catalog.

Teaching Experience

  • Aug. - Oct. 2021
    Co-Instructor - Advanced Machine Learning (Industry Course)
    Vistra Corporation / University of Texas at Dallas
    Co-designed and co-taught an eight-week intensive ML course for engineers at Vistra Corporation, with Prof. Sriraam Natarajan and Dr. Harsha Kokel. The curriculum covered core ML algorithms, probabilistic reasoning, and neural networks, with emphasis on motivation, transfer, and deployment in high-stakes settings. Teaching working professionals sharpened my focus on connecting formalism to problems that matter to practitioners.
  • Mar. 2023
    Tutorial Presenter - Tractable Probabilistic Models
    Indian Institute of Technology, Madras
    Delivered a three-day tutorial on tractable probabilistic models to researchers and ML practitioners, co-presented with Prof. Natarajan, Dr. Saurabh Mathur, and Sahil Sidheekh. With a technically advanced audience and dense material, the focus was on building a coherent conceptual framework and bridging to adjacent areas of specialization.
  • Aug. '20 - May '26
    Graduate Research Assistant and Mentor
    Starling Lab, UT Dallas
    Mentored undergraduate and graduate students in reading technical literature, forming research questions, debugging implementations, and presenting results.

Presentations and Invited Talks

  • 2024
    Tutorial: Deep Tractable Probabilistic Models
    CODS-COMAD 2024
    Co-presented with S. Mathur, S. Sidheekh, and S. Natarajan.
  • Mar. 2023
    Tutorial: Tractable Probabilistic Models
    Indian Institute of Technology, Madras
    Co-presented with S. Natarajan, S. Mathur, and S. Sidheekh.

Academic Service

  • Program Committee / Reviewer
    • AAAI (2022-2027)
    • NeurIPS (2023-2026)
    • ICLR (2024)
  • Workshop Reviewer
    • Neuro Causal and Symbolic AI Workshop at NeurIPS (2022)
    • Tractable Probabilistic Models Workshop at UAI
  • Journal Reviewer
    • Data Mining and Knowledge Discovery
    • Journal of Artificial Intelligence Research

Prior Industry Experience

  • 2015 - 2017
    Software Developer
    NetCracker Technology Pvt. Ltd.
    Developed features for ICOMS (Integrated Communications Operations Management System) across multiple product releases.