Menu

Research

Physics for AI, AI for Physics

Research Interests

Physics for AI

Physics Data for Understanding ML

Leveraging the unique properties of physics datasets (access to controllable data generation and known data symmetry and structure) to probe the internal mechanisms of deep neural networks.

Flow matching comparison
Figure from Generative Models on Phase Space
Physics for AI

Physics-Inspired Scaling Laws

Using an effective field theory to predict how neural network ensembles behave, deriving scaling laws for uncertainty quantification without the need to train an ensemble.

Scaling law predictions
Figure from Uncertainty Quantification from Ensemble Variance Scaling Laws in Deep Neural Networks
AI for Physics

Automating Scientific Model Building

Developing ML methods with the goal of “theory inversion”. Parameter estimation with Uncertainty Quantification, Simulation Emulation, and Automating Theory Writing.

Theory inversion via 2D density estimation
Figure from Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation
AI for Physics

Cross-Domain Foundation Models

Building foundation models for scientific point clouds (irregular graphs!) that transfer knowledge across the domains of particle physics, cosmology, and molecular dynamics.

OmniMol energy predictions
Figure from OmniMol: Transferring Particle Physics Knowledge to Molecular Dynamics with Point-Edge Transformers

All Publications

  1. FAIR Universe Weak Lensing ML Uncertainty Challenge: Handling Uncertainties and Distribution Shifts for Precision Cosmology
    Biwei Dai, et al.
    Accepted to NeurIPS 2026
    arXiv
  2. Uncertainty Quantification from Ensemble Variance Scaling Laws in Deep Neural Networks
    Ibrahim Elsharkawy, et al.
    Machine Learning: Science and Technology, vol. 6, no. 3, 2025
    Paper
  3. Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation
    Ibrahim Elsharkawy and Yonatan Kahn
    arXiv preprint, 2025
    arXiv
  4. Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives
    Ibrahim Elsharkawy, et al.
    arXiv preprint, 2026
    arXiv
  5. OmniMol: Transferring Particle Physics Knowledge to Molecular Dynamics with Point-Edge Transformers
    Ibrahim Elsharkawy, et al.
    arXiv preprint, 2026
    arXiv
  6. Generative Models on Phase Space
    Zachary Bogorad, Ibrahim Elsharkawy, et al.
    arXiv preprint, 2026
    arXiv
  7. FAIR Universe HiggsML Uncertainty Dataset and Competition
    Wahid Bhimji, et al.
    NeurIPS 2025, Dataset and Competition Track
  8. OmniCosmos: Transferring Particle Physics Knowledge across the Cosmos
    Vinicius Mikuni, et al.
    arXiv preprint, 2025
    arXiv

Select Talks & Posters

  • Invited Talk
    Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives
    Vector Institute & Stanford Data Science Seminars, 2026
  • Invited Talk
    1st Place Competition Milestone (Ensembles and Uncertainty Quantification)
    The Challenge of Handling Uncertainties in Fundamental Science @ NeurIPS 2024
  • Invited Talk
    Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation
    CERN 7th Inter-Experimental LHC Machine Learning Workshop, 2025
  • Invited Talk
    Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation
    MIT IAIFI Seminar, 2025
  • Invited Talk
    Contrastive Normalizing Flows for Uncertainty-Aware Parameter Estimation
    ExxonMobil Research Seminar, 2025
  • Poster
    FAIR Universe HiggsML Uncertainty Dataset and Competition
    NeurIPS 2025, Dataset and Competition Track
  • Poster
    Uncertainty Quantification from Scaling Laws in Deep Neural Networks
    Machine Learning and the Physical Sciences Workshop @ NeurIPS 2024
  • Poster
    Pre-Training For Science: A study on Foundation Model Training Objectives
    Stanford Center of Decoding the Universe Forum, 2025

Interested in collaborating?

Contact Me