Physics-Informed Machine Learning (PIML) Resources

A collection of Physics-Informed Machine Learning resources, literature review, tutorials, and code examples. This repository serves as a knowledge hub for researchers, practitioners, and students interested in combining physics knowledge with machine learning.

What is PIML? ☜(ﾟヮﾟ☜)

Physics-Informed Machine Learning (PIML) integrates domain knowledge from physics, engineering, and science into machine learning models. By incorporating physical laws, governing equations, and constraints, PIML methods can:

• Learn from smaller datasets
• Produce physically consistent predictions
• Generalise better to unseen scenarios
• Provide interpretable and trustworthy results

Repository Structure

piml/
├── docs/                      # Documentation and written resources
│   ├── literature-review/     # Comprehensive literature matrices
│   └── fundamentals/          # Core concepts and introductions
├── notebooks/                 # Interactive Jupyter notebooks
└── examples/                  # Standalone code examples
└── resources/                 # Additional materials and datasets

Quick Navigation

• Literature Review contains 50+ papers organized into five thematic matrices

  • Core PIML Papers: Foundational papers on PINNs, neural operators, and domain decomposition
  • Image/Vision Integration: Methods for extracting physics from images (critical for remote sensing)
  • Oceanography & Bathymetry: Wave modeling, bathymetry inversion, and ocean physics
  • Bayesian Methods: Gaussian processes, uncertainty quantification, and probabilistic approaches
  • Software & Tools:Frameworks, libraries, and tutorial resources

• Fundamentals: Introduction to PIML concepts

• Notebooks: Interactive tutorials with explanations

• Examples: Ready-to-use code implementations

Getting Started ฅ(^•ﻌ•^ฅ)

Installation

1. Clone this repository:

git clone https://github.com/yourusername/piml.git
cd piml

2. Install dependencies:

pip install -r requirements.txt

3. Explore the notebooks:

jupyter notebook notebooks/

Contributing

This repository is actively evolving. Contributions, suggestions, and feedback are welcome! Please feel free to:

• Open issues for questions or suggestions
• Submit pull requests for improvements
• Share your own PIML examples or case studies

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Note: This repository consolidates knowledge from the broader PIML research community.

Last Updated: November 2025 | Status: Active Development