ZKP-Authenticated Federated Few-Shot Learning for Secure Client Participation in Distributed Monitoring Devices
Redwanul Karim, Nisha L. Raichur, Lucas Heublein, Tobias Feigl, Christopher Mutschler, Felix Ott
Hi there! I'm Redwanul Karim — part researcher, part engineer, and full-time explorer in the ever-evolving world of Machine Learning. Right now, I'm pursuing my M.Sc. in Artificial Intelligence at FAU Erlangen-Nürnberg , building on the foundations of my B.Sc. in Computer Science & Engineering from North South University , Bangladesh. I spend most of my time trying to make AI smarter, safer, and a little more trustworthy.
At Fraunhofer IIS , I'm researching machine learning security for critical infrastructures like GNSS monitoring, while at the Pattern Recognition Lab , I'm exploring how physics-informed deep learning and large language models can be used to reason over complex, structured knowledge.
Before stepping into academia, I spent three and half exciting years as a Software Engineer at Samsung Research , developing software for Galaxy devices. That experience taught me how to engineer at scale, but my curiosity kept pulling me toward research — diving deep into machine learning security, model interpretability, efficient ML, and reinforcement learning.
Looking forward, I'm actively seeking a Ph.D. position where I can work on secure & trustworthy AI, LLM reasoning, and interpretable ML. My goal is simple — to create machine learning systems that people can trust, understand, and scale safely.
M.Sc. Student
B.Sc. Graduate
Software Engineer
Research Assistant
Student Assistant
Research InternOur paper titled "Physics-informed GNN for medium-high voltage AC power flow with edge-aware attention and line search correction operator" has been accepted to ICASSP 2026.
I have started working as a Student Assistant at the Pattern Recognition Lab (FAU).
I have started working as a Student Assistant in the Robotics Lab at Fraunhofer IIS, Nürnberg, Germany.
I was awarded Global Korea Scholarship (GKS) 2023 to pursue M.Sc. in Artificial Intelligence at UNIST. Candidate Number: CS01230281.
Redwanul Karim, Nisha L. Raichur, Lucas Heublein, Tobias Feigl, Christopher Mutschler, Felix Ott
Redwanul Karim, Changhun Kim, Timon Conrad, Nora Gourmelon, Julian Oelhaf, David Riebesel, Tomás Arias-Vergara, Andreas Maier, Johann Jäger, Siming Bayer
Changhun Kim, Timon Conrad, Redwanul Karim, Julian Oelhaf, David Riebesel, Tomás Arias-Vergara, Andreas Maier, Johann Jäger, Siming Bayer
Redwanul Karim, M.A. Muhaiminul Islam, Sazid Rahman, Saif Ahmed Chowdhury, Kalyan Roy, Adnan Al Neon, Md.Sajid Hasan, Adnan Firoze, Rashedur M. Rahman
Trained LLMs to write shorter, causally-used rationales by rewarding step-level SHAP attribution during PPO.
Trained deep learning models (SchNet, FieldSchNet, PaiNN, SO(3)Net) on MD17 and QM7‑x datasets to predict molecular energies and forces.
Built an ETL data pipeline using advanced data engineering methodologies to correlate solar activity with climate metrics.
March 22, 2026
A rigorous yet accessible presentation of the modern modal ontological argument for the existence of God, examining why simple parody objections fail, where the strongest counterarguments bite, and what the debate ultimately reveals about the limits of pure reason.
Read more