I'm a Postdoctoral Scholar at the University of Southern California, where I also completed my PhD in Engineering (Computational Science) and Master's degrees in Computer Science and Statistics.

This background provides me with a strong foundation in applied statistics, probability theory, and linear algebra, which I leverage in my work at the exciting intersection of Machine Learning, Uncertainty Quantification (UQ), and Scientific Computing.

I specialize in discovering the underlying geometry in complex, high-dimensional data (manifold learning) to build more efficient algorithms for probabilistic modeling and simulation. My passion lies in leveraging AI to accelerate scientific discovery, empower researchers, and unlock new capabilities for the betterment of humanity. Throughout my research, I've had the privilege of collaborating closely with industry leaders like General Motors and General Electric on multiple research projects.

• Actively exploring the foundations of Large Language Models (LLMs), Vision Language Models (VLMs), Natural Language Processing (NLP), and Reinforcement Learning (RL).
• Integrating these advanced AI techniques into novel scientific workflows to automate research, generate hypotheses, and accelerate discovery.

• Accelerating Scientific Design: Leveraging manifold learning and diffusion-based sampling to speed up material design and optimize manufacturing processes.
• Improving Probabilistic Models: Enhancing my open-source Conditional Kernel Density Estimation (CKDE) models for more robust and accurate predictions.

I’m always open to collaborating on open-source projects in scientific computing, ML for science, or uncertainty quantification. If you're working on something interesting in these areas, please feel free to reach out!

• Contact me at: hawi@usc.edu, philippe.hawi@outlook.com, or on LinkedIn

• PLoM: A Python toolkit for Projection on Local Manifolds, designed for efficient probabilistic modeling and generation of high-dimensional data.

• CKDE: An open-source package implementing Conditional Kernel Density Estimation for complex data distributions.

• Analytical certification of engineered systems accounting for data paucity and model error.
  P. Hawi, Z. Yao, V. Aitharaju, J. Mahishi & R. Ghanem.
  Submitted to Reliability Engineering & System Safety (RESS), 2025.


• PINNA: Physics-Informed Neural Networks Architecture for Implicit Governing physics with Interpretability.
  Z. Yao, P. Hawi, V. Aitharaju, J. Mahishi & R. Ghanem.
  Submitted to International Journal for Numerical Methods in Engineering (IJNME), 2025.


• Mesh refinement as probabilistic learning.
  P. Hawi & R. Ghanem.
  Journal of Machine Learning for Modeling and Computing, 2024.


• Stochastic multiscale modeling for quantifying statistical and model errors with application to composite materials.
  Z. Wang, P. Hawi, S. Masri, V. Aitharaju & R. Ghanem.
  Reliability Engineering & System Safety (RESS), 2023.


• Data-driven discovery of free-form governing differential equations.
  S. Atkinson, W. Subber, L. Wang, G. Khan, P. Hawi & R. Ghanem.
  NeurIPS: Second Workshop on Machine Learning and the Physical Sciences, 2019.