PhD Student in Scientific Machine Learning | Johns Hopkins University
I am a PhD student at Johns Hopkins University specializing in Scientific Machine Learning with a focus on developing scalable neural architectures for solving complex physical systems and addressing inverse and control problems. My research bridges the gap between machine learning and computational physics, with applications ranging from biomedical modeling to large-scale PDE solutions.
🔬 Current Research Projects
- Traumatic Brain Injury Modeling: Large-scale brain tissue simulation using neural operators
- High-Dimensional PDE Solving: Scalable algorithms for complex partial differential equations
- Inverse Operator Learning: Novel architectures for system parameter identification
- Differential Predictive Control: DPC for PDE control problems
- Reinforcement learning: Reinforcement leanrning for high dimensional control of PDE problems
Scientific Computing & ML
- Neural Operators (DeepONet, FNO, Graph Neural Networks)
- Physics-Informed Neural Networks (PINNs)
- Multi-GPU and Multi-Node Distributed Computing
- Uncertainty Quantification & Optimization
- Reinforcement Learning for Control
- Differential predictive controls
Programming & Tools
- Languages: Python, MATLAB, Java
- ML Frameworks: Jax, PyTorch, scikit-learn
- Scientific Computing: NumPy, SciPy, FEniCS, deal.ii
- Visualization: Matplotlib, Visit
- Developing scalable neural operator architectures for real-world engineering applications
- Contributing to the intersection of AI and computational physics
- Building robust uncertainty quantification frameworks for scientific applications
- Advancing control theory through physics-informed reinforcement learning
-
🏆 3rd Place - NASA & DNV Challenge on Optimization Under Uncertainty (2025)
- Utilized LightGBM ensemble for uncertainty quantification
- Implemented adaptive differential evolution for robust optimization
-
Mentoring 3 students (undergrad, graduate and high school) in operator learning, PINNs and RAG based agent development.
I'm passionate about pushing the boundaries of AI in scientific computing and would love to discuss research collaborations, open source contributions, or opportunities in applied AI research.
📧 Email: droysar1@jh.edu
🏛️ Institution: Johns Hopkins University
🔬 Research Group: Centrum IntelliPhysics
"Developing neural operators that make solving complex physics as scalable as processing language—enabling real-time simulation of everything from brain injuries to spacecraft dynamics."