About Me
I am a PhD student at Rice University working broadly in machine learning and artificial intelligence, with interests spanning the mathematical foundations of deep learning, generative modeling, and scientific machine learning. Much of my current applied research focuses on problems in the physical sciences, particularly atomistic modeling and materials science.
My work is highly interdisciplinary, and a major part of my research involves bridging the technical languages of machine learning and the physical sciences. I frequently enter new scientific domains, develop the background needed to understand their core modeling problems and constraints, and work with domain experts to translate these into machine-learning formulations. In the other direction, I communicate ML concepts and modeling decisions to scientific collaborators without requiring specialized ML backgrounds. This has included developing presentations and tutorials on topics such as attention and Transformers, generative modeling, optimization, and the physical foundations needed to formulate scientific machine-learning problems.
I am advised by Dr. Anastasios Kyrillidis in Computer Science and Dr. Geoffroy Hautier in Materials Science. I also collaborate with Dr. Christopher Jermaine and Dr. Thomas Reps of UW–Madison. Earlier in my PhD, I worked on machine-learning methods for structural biology and protein crystallography in collaboration with the lab of Dr. George Phillips (ret.).
My research has focused on three main areas:
- Theory and mathematical foundations of neural networks
- Generative modeling, particularly diffusion models
- Scientific machine learning, including machine-learning interatomic potentials (MLIPs) and atomistic modeling
A central theme of my work is understanding how mathematical and problem-specific structure can be incorporated into learning systems while retaining the flexibility and scalability of modern deep learning. My research spans theoretical analysis, model and algorithm development, large-scale experimentation, and the development of research software for practical scientific workflows.
For an up-to-date resume including internships, awards, etc, please feel free to contact me!
Tutorials and Lecture Notes
Diffusion and Related
Transformers and Related
- Lecture Slides on Introduction to Transformers (pdf)
- Reinforcement Learning from Human Feedback Introduction (pdf)
- Transformer Mathematics In Extensive Detail (pdf)
- Deep Dive Into Attention Computations (pdf)