Sharmin Afrose is a R&D associate in Discrete Algorithms Group, Computer Science and Mathematics Division at Oak Ridge National Laboratory (ORNL), Oak Ridge, Tennessee, USA. She obtained her Ph.D. degree in Computer Science at Virginia Tech and her BS degree in Computer Science and Engineering from Bangladesh University of Engineering and Technology.
Her current research focuses on developing trustworthy deep learning algorithms to accelerate scientific discovery, building robust LLM-based quantum compilers and agentic quantum programming systems that make quantum computing more accessible and reliable. She also focus on developing fair and explainable healthcare AI models that improve clinical decision support for high-risk and underserved populations. Previously, she also worked on privacy-preserving federated learning systems using dynamic sketching techniques to reduce communication overhead at scale.
Links
Publications
Mar, 2026
Conference Paper
Dynamical Sketching for Enhanced Communication Efficiency in Federated Learning
Mar, 2026
Conference Paper
Learning-Based Quantum Compilation: Translating QASM to QIR with CodeBERT