Deep Symbolic Regression in Gravitational Wave Foundation Models
Ongoing 6/1/2025
Technologies:
Deep LearningTransformersSymbolic RegressionPythonGravitational Wave Analysis
This project is conducted as a research internship at Radboud University in the High Energy Physics group under the supervision of Prof. Sascha Caron, Nijmegen, Netherlands, from June 2025 to the present.
Project Overview
This project focuses on conducting research on foundation models for gravitational wave analysis using deep transformer-based symbolic regression. I am exploring state-of-the-art methods, including Multi-Modal Symbolic Regression (MMSR), for interpretable modeling of gravitational wave signals.
The work combines advanced machine learning techniques with theoretical physics to develop better methods for analyzing gravitational wave data from detectors like LIGO and Virgo.
Research Goals
- Develop foundation models capable of analyzing complex gravitational wave signals
- Implement transformer-based symbolic regression for interpretable results
- Create models that can discover underlying mathematical relationships in gravitational wave data
- Advance the field of AI for scientific discovery in theoretical physics
Key Technologies
- Deep Learning: Advanced neural network architectures
- Transformers: State-of-the-art attention mechanisms for sequence modeling
- Symbolic Regression: Discovering mathematical expressions from data
- Python: Primary programming language for implementation
- Gravitational Wave Analysis: Domain-specific signal processing techniques