Deepika Vemuri

me.jpeg

I am a PhD student from the AI department at IIT Hyderabad. I am part of the Machine Learning and Vision Group supervised by Dr. Vineeth N Balasubramanian where, broadly, I explore how concept-based learning can help models learn tasks better with a focus, also, on the interpretability of these models. Prior to this, I was an M.Tech student in the University of Hyderabad where I did my thesis under the supervision of Dr. Satish Narayana Srirama on distributed machine learning.

News

Nov 2025 Paper on “LogicCBMs: Using Logic to Enhance Concept-Based Learning” accepted at WACV 2026!
Nov 2025 Selected to participate in the AAAI 2026 Doctoral Consortium!
Aug 2025 Selected for the Vector Internship! Will be hosted at UBC in Feb 2026.
Mar 2025 Invited to the PMRF Annual Symposium at IIT Hyderabad.
Dec 2024 Paper on “Walking the Web of Concept-Class Relationships in Incrementally Trained Interpretable Models” accepted at AAAI 2025!
Feb 2024 Presented our work on “Advancing Ante Hoc Explainable Models through Generative Adversarial Networks” at the Deployable AI and HCLR workshops at AAAI 2024, Vancouver, Canada.
Dec 2022 Received the Prime Minister’s Research Fellowship!

Selected Publications

  1. foca_2.png
    Formal Concept-Based Models
    Deepika Vemuri, Sayanta Adhikari, Ankit Saha, and Vineeth N Balasubramanian
    2026
    Under Review at an A* conference
  2. logiccbms.png
    LogicCBMs: Using Logic to Enhance Concept-Based Learning
    Deepika Vemuri, Gautham Bellamkonda, Aditya Pola, and Vineeth N Balasubramanian
    2026
    Winter Conference on Applications of Computer Vision (WACV)
  3. foca3.png
    Learning Hierarchically Using Formal Concepts
    Deepika Vemuri, Sayanta Adhikari, Ankit Saha, and Vineeth N Balasubramanian
    In Visual Concepts Workshop, CVPR, 2025
  4. concept_relations.png
    Walking the Web of Concept-Class Relationships in Incrementally Trained Interpretable Models
    Susmit Agrawal, Deepika Vemuri, Sri Siddarth Chakaravarthy P, and Vineeth N Balasubramanian
    In AAAI, 2025
  5. logic_w_2.png
    Enhancing Concept-based Learning with Logic
    Deepika Vemuri, Gautham Bellamkonda, and Vineeth N Balasubramanian
    In Differentiability and AI Safety Workshops at ICML, 2024
  6. advancing_antehoc.png
    Advancing Ante Hoc Explainable Models through Generative Adversarial Networks
    Tanmay Garg, Deepika Vemuri, and Vineeth N Balasubramanian
    In Deployable AI Workshop, HCLR Workshop, AAAI, 2024
  7. canto2.png
    CANTO: An Actor Model-based Distributed Fog Framework Supporting Neural Networks Training in IoT Applications
    Satish Narayana Srirama and Deepika Vemuri
    Journal of Computer Communications, 2023