I’m Subrat Kishore Dutta, a first year PhD student at
CISPA Helmholtz Center for Information Security and
University of Hamburg under the supervision of
Dr. Xiao Zhang and
Prof. Dr. Anne Lauscher.
My research centres around Adversarial Machine Learning and its utility in understanding and improving
Fairness of Foundational Models. I completed my MSc. in Informatik at the
Universität des Saarlandes,
where I published my Master's Thesis at ICCV 2025.
Research Motivation and Interests
I believe that representation plays a critical role in ensuring fairness across individuals from diverse backgrounds. Real-world data often reflects societal biases, which are then inherited and amplified by AI systems. As a result, marginalized or smaller subgroups are disproportionately at risk. In my research, I intent to explore methods that can improve both the fairness and representational capacity of machine learning systems.
- Adversarial Machine Learning
- Fairness and Bias in Foundational Models
- Interpretability
TA Experience
- Algorithmic Foundations of Adversarial Robustness (Spring 2025-UHH)
Miscellaneous
- ⚙️ During my bachelors I enjoyed arduino programming!
- 👁️ I like Computer Vision application problems.
- 🎵 I am into Indian classical music.
- 🥊 I tried boxing during my masters!
- 📷 I sometimes take photos.
Research Experience
PhD Student
May, 2025 — Present
Research Assistant
Jan, 2024 — April, 2025
Education
